an empirical assessment of country risk ratings and ... · inoue, 1985). country risk assessment...

UNCORRECTE

DPROOF

AN EMPIRICAL ASSESSMENT OFCOUNTRY RISK RATINGS AND

ASSOCIATED MODELS

Suhejla Hoti and Michael McAleer

University of Western Australia

Abstract. Country risk has become a topic of major concern for the internationalfinancial community over the last two decades. The importance of country ratingsis underscored by the existence of several major country risk rating agencies,namely the Economist Intelligence Unit, Euromoney, Institutional Investor, Inter-national Country Risk Guide, Moody’s, Political Risk Services, and Standard andPoor’s. These risk rating agencies employ different methods to determine countryrisk ratings, combining a range of qualitative and quantitative information regard-ing alternative measures of economic, financial and political risk into associatedcomposite risk ratings. However, the accuracy of any risk rating agency with regardto any or all of these measures is open to question. For this reason, it is necessaryto review the literature relating to empirical country risk models according toestablished statistical and econometric criteria used in estimation, evaluation andforecasting. Such an evaluation permits a critical assessment of the relevance andpracticality of the country risk literature. The paper also provides an internationalcomparison of risk ratings for twelve countries from six geographic regions. Theseratings are compiled by the International Country Risk Guide, which is the onlyrating agency to provide detailed and consistent monthly data over an extendedperiod for a large number of countries. The time series data permit a comparativeassessment of the international country risk ratings, and highlight the importanceof economic, financial and political risk ratings as components of a compositerisk rating.

Keywords. Country risk; Economic risk; Financial risk; Political risk; Compositerisk; Risk ratings; Risk returns; Volatilities; Component analysis; Internationalcomparison.

1. Introduction

1.1. Country risk

Following the rapid growth in the international debt of less developed countriesin the 1970s and the increasing incidence of debt rescheduling in the early 1980s,country risk, which reflects the ability and willingness of a country to service itsfinancial obligations, has become a topic of major concern for the internationalfinancial community (Cosset and Roy, 1991). Political changes resulting from the

0950-0804/04/00 0001–50 JOURNAL OF ECONOMIC SURVEYS Vol. 18, No. 0# Blackwell Publishing Ltd. 2004, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main St.,Malden, MA 02148, USA.

CSE: AR JOES 001

UNCORRECTE

DPROOF

fall of communism, and the implementation of market-oriented economic andfinancial reforms, have resulted in an enormous amount of external capitalflowing into the emerging markets of Eastern Europe, Latin America, Asia, andAfrica (Ramcharran, 1999). These events have alerted international investors tothe fact that the globalisation of world trade and open capital markets are riskyelements that can cause financial crises with rapid contagion effects, whichthreaten the stability of the international financial sector (Hayes, 1998). In lightof the tumultuous events flowing from September 11, 2001, the risks associatedwith engaging in international relationships have increased substantially, andbecome more difficult to analyse and predict for decision makers in the economic,financial and political sectors.

Given these new developments, the need for a detailed assessment of countryrisk and its impact on international business operations is crucial. Country riskrefers broadly to the likelihood that a sovereign state or borrower from aparticular country may be unable and/or unwilling to fulfil their obligationstowards one or more foreign lenders and/or investors (Krayenbuehl, 1985). Aprimary function of country risk assessment is to anticipate the possibility of debtrepudiation, default or delays in payment by sovereign borrowers (Burton andInoue, 1985). Country risk assessment evaluates economic, financial, and politicalfactors, and their interactions in determining the risk associated with a particularcountry. Perceptions of the determinants of country risk are important becausethey affect both the supply and cost of international capital flows (Brewer andRivoli, 1990).

Country risk may be prompted by a number of country-specific factors orevents. There are three major components of country risk, namely economic,financial and political risk. The country risk literature holds that economic,financial and political risks affect each other. As Overholt (1982) argues, inter-national business scenarios are generally political-economic as businesses andindividuals are interested in the economic consequences of political decisions.

The lending risk exposure vis-a-vis a sovereign government is known as sovereignrisk (Juttner, 1995). According to Ghose (1988), sovereign risk emerges when asovereign government repudiates its overseas obligations, and when a sovereigngovernment prevents its subject corporations and/or individuals from fulfillingsuch obligations. In particular, sovereign risk carries the connotation that therepudiation occurs in situations where the country is in a financial position tomeet its obligations. However, sovereign risk also emerges where countries areexperiencing genuine difficulties in meeting their obligations. In an attempt toextract concessions from their lenders and to improve rescheduling terms, negoti-ators sometimes threaten to repudiate their ‘borrowings’ (Bourke, 1990).

Political risk is generally viewed as a non-business risk introduced strictlyby political forces. Banks and other multinational corporations have identifiedpolitical risk as a factor that could seriously affect the profitability of theirinternational ventures (Shanmugam, 1990). Ghose (1988) argues that politicalrisk is analogous to sovereign risk and lies within the broader framework ofcountry risk. Political risk emerges from events such as wars, internal and external

2 HOTI AND MCALEER

# Blackwell Publishing Ltd. 2004

UNCORRECTE

DPROOF

conflicts, territorial disputes, revolutions leading to changes of government, andterrorist attacks around the world. Social factors include civil unrests due toideological differences, unequal income distribution, and religious clashes.Shanmugam (1990) introduces external reasons as a further political aspect ofcountry risk. For instance, if the borrowing nation is situated alongside a countrythat is at war, the country risk level of the prospective borrower will be higherthan if its neighbour were at peace. Although the borrowing nation may not bedirectly involved in the conflict, the chances of a spillover effects may exist.Additionally, the inflow of refugees from the war would affect the economicconditions in the borrowing nation. In practical terms, political risk relates tothe possibility that the sovereign government may impose foreign exchange andcapital controls, additional taxes, and asset freezes or expropriations. Delays inthe transfer of funds can have serious consequences for investment returns,import payments and export receipts, all of which may lead to a removal of theforward cover (Juttner, 1995).

Economic and financial risks are also major components of country risk. Theyinclude factors such as sudden deterioration in the country’s terms of trade, rapidincreases in production costs and/or energy prices, unproductively investedforeign funds, and unwise lending by foreign banks (Nagy, 1988). Changes inthe economic and financial management of the country are also importantfactors. These risk factors interfere with the free flow of capital or arbitrarilyalter the expected risk-return features for investment. Foreign direct investors arealso concerned about disruptions to production, damage to installations, andthreats to personnel (Juttner, 1995).

1.2. Country risk ratings

Since the Third World debt crisis in the early 1980s, commercial agencies such asMoody’s, Standard and Poor’s, Euromoney, Institutional Investor, EconomistIntelligence Unit, International Country Risk Guide, and Political Risk Services,have compiled sovereign indexes or ratings as measures of credit risk associatedwith sovereign countries. Risk rating agencies provide qualitative and quantita-tive country risk ratings, combining information about alternative measures ofeconomic, financial and political risk ratings to obtain a composite risk rating.This paper provides an international comparison of country risk ratings andreturns compiled by the International Country Risk Guide (ICRG), which isthe only risk rating agency to provide detailed and consistent monthly dataover an extended period for a large number of countries.

Derivative assets, such as futures and options, are used to hedge against pricerisk in commodity markets. In particular, country risk ratings are used to hedgeagainst issued bonds. Optimal hedging strategies and an evaluation of the riskassociated with risk ratings require knowledge of the volatility of the underlyingprocess. As volatility is generally unknown, it must be estimated. These estimatedand predicted volatilities are fundamental to risk management in financial modelsthat describe the risk-return trade-off. Although there does not yet seem to be an

COUNTRY RISK RATINGS 3


UNCORRECTE

DPROOF

established market for pricing risk ratings as a primary or derivative asset,estimating and testing the volatility associated with risk ratings would seem tobe a first step in this direction. In the finance and financial econometrics litera-ture, conditional volatility has been used to evaluate risk, asymmetric shocks, andleverage effects. The volatility present in risk ratings also reflects risk considera-tions in risk ratings. As risk ratings are effectively indexes, their rate of change (orreturns) merits attention in the same manner as financial returns (for furtherdetails, see Hoti et al., 2002).

The plan of the paper is as follows. Section 2 provides a quantitative classi-fication of empirical country risk models, which forms the database, and alsoclassifies and describes the data. Various theoretical and empirical modelspecifications used in the literature are reviewed analytically and empirically inSection 3. Section 4 discusses the empirical findings of the published studies. Acomparison of ICRG country risk ratings, risk returns, and their associatedvolatilities for twelve representative developing countries is given in Section 5.Concluding remarks and some suggestions for future research are presented inSection 6.

2. Classification of Country Risk Models and the Data

For purposes of evaluating the significance of empirical models of country risk, itis necessary to analyse such models according to established statistical andeconometric criteria. The primary purpose of each of these empirical papers isto evaluate the practicality and relevance of the economic, financial and politicaltheories pertaining to country risk. An examination of the empirical impact andstatistical significance of the results of the country risk models will be based on anevaluation of the descriptive statistics relating to the models, as well as theeconometric procedures used in estimation, testing and forecasting.

This paper reviews 50 published empirical studies on country risk (the papersare listed in the Appendix). A classification of the 50 empirical studies is givenaccording to the model specifications examined, the choice of dependent andexplanatory variables considered, the number of explanatory variables used,econometric issues concerning the recognition, type and number of omittedexplanatory variables, the number and type of proxy variables used when vari-ables are omitted, the method of estimation, and the use of diagnostic tests of theauxiliary assumptions of the models.

Scrutiny of the ECONLIT software package and the Social Science CitationIndex for the most widely cited articles in the Country Risk literature yieldsat least 50 published empirical papers over the last three decades in refereedjournals. Although the first two papers were published by Frank and Cline(1971) (Journal of International Economics) and Feder and Just (1977) (Journalof Development Economics), there were 16 papers published in the 1980s, afurther 30 papers published in the 1990s, and with the 2 most recent papershaving been published in 2001. Thus, the literature is essentially two decadesold. There is no leading journal in the literature on country risk, with the Journal

4 HOTI AND MCALEER


UNCORRECTE

DPROOF

of Development Economics publishing 6 papers, the Journal of InternationalBusiness Studies publishing 4 papers, the Journal of Banking and Finance,Economics Letters and Applied Economics each publishing 3 papers, AppliedEconomics Letters and Global Finance Journal each publishing 2 papers, and 27other journals each publishing one paper on the topic.

Country risk has been surveyed previously by Saini and Bates (1984) and Eatonand Taylor (1986). Saini and Bates (1984) provide a survey of the quantitativeapproaches to risk analysis by reviewing the problems in the statisticalapproaches of published empirical papers. In particular, they examine the short-comings with regard to definitions of dependent variables, the quality and avail-ability of data, model specifications, appropriateness of statistical methods, andthe ability to forecast debt servicing difficulties adequately. Eaton and Taylor(1986) review the theoretical aspects of numerous papers relating to LDC debtand financial crises, with an emphasis on the policy implications to be drawn.Although they do not analyse the empirical aspects of the various papers, theyexamine the three main issues in empirical applications, namely the determinantsof rescheduling, how credit terms are fixed, and the factors determining theamounts borrowed. While the primary purpose of Rockerbie (1993) is to explainthe interest spread on sovereign Eurodollar loans on the basis of various indica-tors of default risk in lesser developed countries and developed countries, heprovides a useful summary of risk indicators in the empirical papers examined.Thus, the present paper may be seen as a continuation of these surveys using morerecently published contributions to the literature on country risk.

In Table 1, the 50 studies are classified according to the type of data used,namely cross-section or pooled, which combines time series and cross-sectionsamples. Common sources of data are the International Monetary Fund, Bankfor International Settlements, various sources of the World Bank, Euromoney,Institutional Investor, Moody’s, Standard and Poor’s, and various country-specificstatistical bureaux. Almost three-quarters of the studies are based on pooled data,with the remaining one-quarter based on cross-section data.

Table 2 classifies the 34 studies using pooled data according to the number ofcountries, which varies from 5 to 95 countries, with mean 48 and median 47, withthe frequency of occurrence of each number generally being 1. The same 34studies using pooled data are classified according to the number of annual andsemi-annual observations in Tables 3 and 4, respectively. For the annual observations,the range of the 19 data sets is 5–24 years, with the mean, median and mode of thenumber of observations being 12, 11 and 5, respectively, with the frequency of

Table 1. Classification by Type of Data Used.

Type of data Frequency

Pooled 34Cross-section 16Total 50



UNCORRECTE

DPROOF

occurrence of each number varying between 1 and 5. The range of the 8 data sets usingsemi-annual observations is 8 to 38 half-years, with themean andmedian andmode ofthe number of observations being 18.5 and 17, respectively.

Tables 5 and 6 classify the studies using cross-section data according to thenumber of countries and the number of time series observations, respectively. InTable 5, 1 study did not report the number of countries used, while another studyused data on 892 municipalities. Of the remaining 16 studies, the range is 18 to143 countries, with mean 55.3 and median 50.5. There are 29 data sets using timeseries observations in Table 6, with range 1 to 23, mean 5.3, median 3, and mode1. Indeed, the most commonly used number of time series observations is 1, with afrequency of 10 in the 29 data sets, so that more than one-third of the cross-section data sets used are based on a single year.

3. Theoretical and Empirical Model Specifications

The general country risk model typically estimated, tested and evaluated is given as:

f Yt;Xt; ut;�ð Þ ¼ 0 ð1Þ

in which f(.) is an unspecified functional form, Y is the designated (vector of)endogenous variables, X is the (vector of) exogenous variables, u is the (vector of)errors, � is the vector of unknown parameters, and t¼ 1, . . . ,n observations. As

Table 2. Classification of Pooled Data by Number of Countries.

Numbers of countriesNumber ofstudies Frequency

5, 16, 17, 19, 24, 25, 26, 30, 32, 39, 41, 43, 48, 54, 55, 56, 60,65, 68, 74, 75, 80, 85, 90, 95 25 127, 33, 40, 47, 59, 79 6 2Total 37

Note: Three studies used two data sets.

Table 3. Classification of Pooled Data by Number of Annual Observations.

Numbers of observations Frequency

5 58, 10, 12, 19 39 211 413, 14, 15, 16, 17, 18, 22, 24 1Total 31

Note: One study used two annual data sets, two studies used one annual data set and one semi-annualdata set, and another study used one annual data set, one semi-annual data set, and one monthly data set.

6 HOTI AND MCALEER


UNCORRECTE

DPROOF

will be discussed below, equation (1) is typically given as a linear or log-linearregression model, or as a logit, probit or discriminant model. The elements of Yand X will also be discussed below. Defining the information set at the end ofperiod t� 1 as It�1¼ [Yt�1,Yt�2,Yt�3, . . . ;Xt,Xt�1,Xt�2,Xt�3, . . . ], the assumptionsof the classical model are typically given as follows:

(A1) E(ut)¼ 0 for all t;(A2) constant unconditional and conditional variances of ut;(A3) serial independence (namely, no covariation between ut and us for t 6¼ s);(A4) X is weakly exogenous (that is, there is no covariation between Xt and us

for all t and s);(A5) u is normally distributed;(A6) parameters are constant;(A7) Y and X are both stationary processes, or are cointegrated if both are

non-stationary.

Diagnostic tests play an important role in modern empirical econometrics, andare used to check the adequacy of a model through testing the underlyingassumptions. The standard diagnostic checks which are used to test assumptions(A1) through (A7) are various tests of functional form misspecification, hetero-scedasticity, serial correlation, exogeneity, third- and higher-order moments ofthe distribution for non-normality, constancy of parameters and structuralchange, unit root tests, and tests of cointegration. There is, in general, little orno theoretical basis in the literature for selecting a particular model of countryrisk. In empirical analysis, however, computational convenience and the ease of

Table 4. Classification of Pooled Data by Number of Semi-Annual Observations.


8, 17, 22 216, 38 1Total 8

Note: One study used two semi-annual data sets, two studies used one annual data set and one semi-annual data set, and another study used one annual data set, one semi-annual data set, and onemonthly data set.

Table 5. Classification of Cross-section Data by Number of Countries.

Numbers of countries Frequency

18, 20, 27, 29, 30, 35, 49, 52, 71, 88, 93, 143, 892, unstated 145, 70 2Total 18

Note: One study used three data sets. The sample with 892 observations refers to municipalities ratherthan countries.



UNCORRECTE

DPROOF

interpretation of models are primary considerations for purposes of modelselection.

Of the 70 models used in the 50 studies, which are reported in Table 7, all butsix are univariate models. The most popular model in the literature is the logitmodel, which is used 23 times, followed by the probit, discriminant, and Tobitmodels, which are used 10, 7, and 3 times, respectively. Thus, more than halfof the models used in the literature are probability-based models. Given thepopularity of the linear and log-linear regression models in empirical economicresearch, it is surprising to see that the linear regression model is used four times,the log-linear regression model is used only twice, and both regression modelsare used in the same study only twice. The artificial neural network model isalso used twice. Of the remainder, the multi-group hierarchical discriminationmodel, two-way error components model, random-effect error component equa-tions, naive model, combination model, G-logit model, nested trinomial logit,

Table 6. Classification of Cross-Section Data by Number of Time Series Observations.


1 102 43, 7, 8, 10, 11, 23 14, 20 25 5Total 29

Note: More than one time series data set was used in some studies.

Table 7. Classification by Type of Model.

Model Frequency

Only linear single equations 4Only log-linear single equations 2Both linear and log-linear single equations 2Logit 23Probit 10Discriminant model 7Tobit 3System of equations 6Artificial neural network model 2Others 11

Total 70

Note: More than one model was used in some studies and two studies used no model. The ‘Others’category includes one entry for each of multi-group hierarchical discrimination model, two-way errorcomponents model, random-effect error component equations, naıve model, combination model,G-Logit model, nested trinomial logit, sequential-response logit, unordered-response logit, classificationand regression trees, and cluster analysis.

8 HOTI AND MCALEER


UNCORRECTE

DPROOF

sequential-response logit, unordered-response logit, classification and regressiontrees, and cluster analysis, are used once each.

The dependent variable for purposes of analysing country risk is broadlyclassified as the ability to repay debt. Of the different types of dependent variablesgiven in Table 8, with more than one dependent variable being used in somestudies, the most frequently used variable is debt rescheduling, which is used 36times. This dependent variable is defined as the probability of general, commer-cial, and official debt rescheduling or debt default (in the current year or in thefuture), and discriminant score of whether a country belongs to a rescheduling ornon-rescheduling group. The second most frequently used variable is agencycountry risk ratings, which is used 18 times. In the empirical analyses, this

Table 8. Classification by Type of Dependent Variable Used1.

Type Frequency

Debt rescheduling2 36Agency country risk ratings3 18Debt arrears4 4(Average) value of debt rescheduling 3Exchange rate movements 3Fundamental valuation ratios 3Demand for debt 3Supply of debt 3Propensity to obtain agency municipality credit risk ratings 2Public debt to private creditors 2Total reserves 2(Relative) bond spreads 2Weighted average loan spread 1Spread over LIBOR 1Yield spreads of international bonds 1Payment interruption likelihood index 1Sovereign loan default 1Credit risk rating 1Income classification 1Stock returns 1Secondary market price of foreign debt 1Dummy for debt crisis 1Total 91

Notes:1. More than one dependent variable was used in some studies.2. Includes variables defined as the probability of debt rescheduling (as proxy for debt default), the

probability of partial reneging when a borrower has decided to reschedule, trichotomous variable ofdebt rescheduling, the probability of general, commercial, official, and band debt rescheduling (inthe current year or in the future), the probability of debt default, and discriminant score of whethera country belongs to a rescheduling or non-rescheduling group.

3. Refers to Institutional Investor, Euromoney, Standard and Poor’s, Moody’s, and EconomistIntelligence Unit country or municipality credit risk ratings, and average agency country riskratings.

4. Includes one entry for each of limit on debt arrears, dummy for significant debt arrears, probabilityof experiencing significant debt arrears, and probability of emerging debt-servicing arrears.



UNCORRECTE

DPROOF

dependent variable is defined as Institutional Investor, Euromoney, Standard andPoor’s, Moody’s, and Economist Intelligence Unit country or municipality riskratings, and the average of agency country risk ratings. Ten types of dependentvariable are used more than once, with debt arrears (defined as the limit ondebt arrears), dummy for significant debt arrears, probability of experiencingsignificant debt arrears, and probability of emerging debt-servicing arrearsbeing used 4 times each, and average value of debt rescheduling, exchange ratemovements, fundamental valuation ratios, demand for debt, and supply of debtbeing used 3 times each. Dependent variables, such as the propensity to obtainagency municipality credit risk ratings, public debt to private creditors, totalreserves, and total or relative bond spread, are used twice each, with the remain-ing 10 types of dependent variable, which are used once each, including weightedaverage loan spread, spread over LIBOR, yield spreads of international bonds,payment interruption likelihood index, sovereign loan default, credit risk rating,income classification, stock returns, secondary market price of foreign debt, anddummy for debt crisis.

There are three types of explanatory variables used in the various empiricalstudies, namely economic, financial and political. Treating country risk variablesas economic and/or financial, and regional differences as political, Tables 9 and10 present the numbers of each type of variable and their frequency. In Table 9,the number of economic and financial variables ranges from 2 to 32, with mean11.5, median 8 and mode 6. Seven of the 19 sets of economic and financialvariables have a frequency of one, with a frequency of 2 occurring 3 times, afrequency of 3 occurring 5 times, and frequencies of 4, 5, and 6 occurring onceeach. In Table 10, the number of political variables ranges from 0 to 13, withmean 1.86, median 0 and mode 0. The absence of any political variable occurs 30times in the 50 studies.

Of the remaining 10 sets of political variables, 2 have a frequency of 4, one hasa frequency of 3, 2 have a frequency of 2, and five have a frequency of one.Hundreds of different economic, financial and political explanatory variableshave been used in the 50 separate studies. The set of economic and financialvariables includes indicators for country risk ratings, debt service, domestic and

Table 9. Classification by Number of Economic and Financial Explanatory Variables.

Number Frequency

2, 3, 7, 13, 16 34 45, 9, 10 26 78 511, 14, 15, 18, 20, 23, 32 112 6Total 50

Note: Country risk indicators are treated as economic and/or financial variables.

10 HOTI AND MCALEER


UNCORRECTE

DPROOF

international economic performance, domestic and international financial perform-ance, monetary reserves, and structural differences. Indicators for countrypolitical risk ratings, domestic and international armed conflict, political events,and regional differences are used in the set of political variables.

The unavailability of the required data means that proxy variables have fre-quently been used for the unobserved variables. Tables 11 and 12 are concernedwith the important issue of omitted explanatory variables in each of the 50studies. It is well known that, in general, omission of relevant explanatory vari-ables from a linear regression model yields biased estimates of the coefficients ofthe included variables, unless the omitted variables are uncorrelated with each ofthe included explanatory variables. For non-linear models, consistency replacesunbiasedness as a desirable statistical characteristic of an estimation method. Insome studies, there is an indication of the various types of variables that arerecognised as being important. Nevertheless, some of these variables have beenomitted because they are simply unavailable. The classification in Table 11 is byrecognition of omitted explanatory variables, where the recognition is explicitlystated in the study. Such an explicit recognition of omitted explanatory variablesis used primarily as a check of consistency against the number of proxy variablesused.

Of the 50 studies in Table 11, exactly three-fifths did not explicitly recognisethat any variables had knowingly been omitted, with the remaining 20 studies

Table 10. Classification by Number of Political Explanatory Variables.

Number Frequency

0 301, 2 43, 8, 10, 11, 13 14, 5 26 3Total 50

Note: Regional differences are treated as political variables.

Table 11. Classification by Recognition of Omitted Explanatory Variables.

Number omitted Frequency

0 301 132, 3, 4 28 1

Total 50

Note: The classification is based on explicit recognition of omitted explanatory variables, and is usedprimarily as a check of consistency against the number of proxy variables used in the correspondingstudies.



UNCORRECTE

DPROOF

recognising that 39 explanatory variables had been omitted. The number ofexplanatory variables explicitly recognised as having been omitted varies from 1to 8. Including and excluding the 30 zero entries for omitted explanatory variablesgive mean numbers omitted of 0.78 and 1.95, respectively, medians of 0 and 1,and modes of 0 and 1. Thirteen of the 20 studies, which explicitly recognised theomission of explanatory variables, noted that a single variable had been omitted.

The classification in Table 12 is given according to the type of omitted explan-atory variable, which is interpreted as predominantly economic and financial orpolitical. More than two-thirds of the omitted explanatory variables are pre-dominantly economic and financial in nature, and the remaining one-third ispredominantly political. Somewhat surprisingly, very few studies stated dynamicsas having been omitted from the analysis, even though most did not explicitlyincorporate dynamics into the empirical specifications.

As important economic, financial and political explanatory variables have beenrecognised as having been omitted from two-fifths of the 50 studies (see Table 11),proxy variables have been used in most of these studies. Tables 13 and 14 areconcerned with the issues of the number and type of proxy variables used. Theproblems associated with the use of ordinary least squares (OLS) to estimate theparameters of linear models in the presence of one or more proxy variables aregenerally well known in the econometrics literature, but extensions to non-linearmodels, which dominate the literature on country risk, are not yet available.Nevertheless, as a guide for analysis, the basic results are outlined below. Theseresults are of special concern as one-half of the studies explicitly recognises theomission of at least one explanatory variable.

In the case where only one proxy variable is used to replace a variable which isunavailable, the well-known results are as follows: (1) the absolute bias in theestimated coefficient of the proxy variable is less than the case where the proxyvariable is excluded; (2) the absolute bias in the estimated coefficient of thecorrectly measured variable is less than in the case where the proxy variable isexcluded; (3) a reduction in measurement error is beneficial; and (4) it is prefer-able to include the proxy variable than to exclude it. When two or more proxyvariables are used to replace two or more variables, which are unavailable, it isnot necessarily the case that the four basic results stated above actually hold.Thus, among other outcomes, the absolute bias in the estimated coefficients ofboth the correctly measured and incorrectly measured variables may be higher if

Table 12. Classification by Type of Omitted Explanatory Variables.

Omitted variable Frequency

Economic and financial factors 28Political factors 11Total 39

Notes: The various omitted variables are classified according to whether they are predominantlyeconomic and financial or political in nature.

12 HOTI AND MCALEER


UNCORRECTE

DPROOF

two or more proxy variables are not used than when they are used, a reduction inmeasurement error may not be beneficial, and it may not be preferable to includetwo or more proxy variables than to exclude them. The reason for the differentoutcomes is that the covariation in two or more measurement errors may exacer-bate the problem of measurement error rather than containing it.

Table 13 classifies the 20 studies by the use of proxy variables, which rangesfrom 1 to 7. Including and excluding the 2 zero entries for the number of proxyvariables used give mean numbers omitted of 2.45 and 2.72, respectively, amedian of 2 in each case, and a mode of 1 in each case. By comparison withTable 11, in which 13 of the 20 studies explicitly recognised the omission of asingle explanatory variable, Table 13 shows that only 7 studies used a single proxyvariable. Otherwise, the results in Tables 11 and 13 are reasonably similar.

The classification in Table 14 is given according to the type of proxy variableused, which is interpreted as comprising predominantly economic and financial orpolitical factors. More than two-thirds of the proxy variables are predominantlyeconomic and financial in nature, and the remaining one-third is predominantlypolitical, which is very similar to the results given in Table 12.

In Table 15 the classification is by method of estimation, in which more than oneestimation method is used in some studies. Five categories are listed, namely OLS,maximum likelihood (ML), Heckman’s two-step procedure, discriminant methods,and Others, which includes entries for, among others, propagation algorithm,regression-based techniques, approximation, minimax, Bayesian, optimal minimumdistance, stepwise, optimisation, binary splits, jack-knife methods and OLS andWLS.Even though logit, probit, and Tobit models in Table 7 are used 40 times in total, ML

Table 14. Classification by Type of Proxy Variables Used.

Proxy variables Frequency

Economic and financial factors 34Political factors 15Total 49

Note: Some studies used economic, financial and political proxy variables.

Table 13. Classification by Number of Proxy Variables Used.

Number Frequency

0, 3, 6 21 72 44, 5, 7 1Total 20

Note: Two studies explicitly recognized the omission of explanatory variables but used no proxyvariables.



UNCORRECTE

DPROOF

is used for estimation purposes only 35 times. Moreover, while linear and log-linearmodels are used only 7 times in total in Table 7, OLS is used 14 times in Table 15 (15times if both OLS andWLS are included). Finally, while discriminant models are used7 times in Table 7, discriminant estimation is used only three times in Table 15.

Finally, the classification in Table 16 is by use of diagnostics to test one or moreauxiliary assumptions of the models. The role of diagnostic tests has become wellestablished in the econometrics literature in recent years, and plays an increasinglyprominent role in modern applied econometrics (see McAleer (1994) for furtherdetails). Most diagnostic tests of the auxiliary assumptions are standard, and areavailable in widely used econometric software packages. Unbelievably, 42 of the 50studies did not report any diagnostic tests whatsoever. Of the eight which did reportany diagnostic tests at all, there were two entries for White’s standard errors forheteroscedasticity, and one entry for each of WLS and heteroscedasticity, transfor-mation for non-normality, White’s covariance matrix for heteroscedasticity, Chowtest, Hajivassiliou’s test for exogeneity, and serial correlation. This is of seriousconcern, especially as the ML method is known to lack robustness to departuresfrom the stated assumptions, but is nevertheless used 35 times. Models such as thelogit and probit are also sensitive to departures from the underlying logistic andnormal densities, respectively, so that the underlying assumptions should be checkedrigorously. As the use of diagnostics has been ignored in the country risk literature, ingeneral, the empirical results should be interpreted with some caution and scepticism.

Table 15. Classification by Method of Estimation.

Method Frequency

OLS 14ML 35Heckman’s two-step procedure 2Discriminant methods 3Others 17Total 71

Note: More than one estimation method was used in some studies. The ‘Others’ category includesentries for, among others, propagation algorithm, regression-based technique, approximation, mini-max, Bayesian, optimal minimum distance, stepwise optimisation, binary splits, jack-knife methods,and OLS and WLS.

Table 16. Classification by Use of Diagnostics.

Type of diagnostics Frequencies

None 42Others 8Total 50

Note: The ‘Others’ category includes entries for WLS and heteroscedasticity, White’s standard errorsfor heteroscedasticity, White’s covariance matrix for heteroscedasticity, Chow test, transformation fornon-normality, Hajivassiliou test for exogeneity, and serial correlation.

14 HOTI AND MCALEER


UNCORRECTE

DPROOF

4. Empirical Findings for Country Risk Ratings

Of the 91 types of dependent variables used in the 50 studies, 27 studies examineddebt rescheduling on 36 occasions and 17 considered country risk ratings on 18occasions (see Table 8 for definitions of these two types of variables). Table 17reports four types of risk component variables used in the 17 country risk ratingsstudies, namely economic, financial, political, and composite. Composite riskvariables are ratings or aggregates that comprise economic, financial and politicalrisk component variables, and were used in all 17 studies. Of these studies, onlytwo did not use economic variables and only one did not use financial variables.Political variables have been used less frequently, namely in 10 studies. Table 18presents the number of country risk components used, as well as their frequency.All four country risk components have been used in 10 studies, 4 studies usedvariables representing three risk components, 3 studies used variables represent-ing two risk components, and no study used variables representing only one riskcomponent.

In Table 19, the 17 are classified according to the risk rating agency they used,namely Institutional Investor, Euromoney, Moody’s Standard and Poor’s, Inter-national Country Risk Guide, Economist Intelligence Unit, and Political RiskServices. These agencies are leading commercial analysts of country risk. Whilethe rating system for the International Country Risk Guide will be analysed in thenext section, the rating systems for the other agencies are briefly discussed below.Unless otherwise stated, the information regarding the agency rating systems hasbeen obtained from the website of Foreign Investment Advisory Service Program,which is a joint service of two leading multilateral development institutions,namely the International Finance Corporation and World Bank (http://www.fias.net/investment_ climate.html).

Institutional Investor compiles semi-annual country risk surveys, which arebased on responses provided by leading international banks. Bankers from 75to 100 banks rate more than 135 countries on a scale of 0 to 100, with 100representing the lowest risk. The individual ratings are weighted using the Institu-tional Investor formula, with greater weights assigned to responses based on theextent of a bank’s worldwide exposure and the degree of sophistication of abank’s country risk model. The names of the participating banks are keptstrictly confidential (Howell, 2001). Institutional Investor country risk surveysare published in the March and September issues of the monthly magazine. In the

Table 17. Risk Component Variables Used in Country Risk Ratings.

Variables Frequency

Economic 15Financial 16Political 10Composite 17Number of studies 17



UNCORRECTE

DPROOF

country risk literature, the Institutional Investor country risk assessment isknown as the banker’s judgment.

Like Institutional Investor, Euromoney provides semi-annual risk ratings andrankings for 185 sovereign countries. Countries are given their respective scoresbased on nine components, and are ranked accordingly. In order to obtain theoverall country risk score, a weight is assigned to each of the nine categories(political risk, 25%; economic performance, 25%; debt indicators, 10%; debt indefault or rescheduled, 10%; credit ratings, 10%; access to bank finance, 5%;access to short-term finance, 5%; access to capital markets, 5%; and discount onforfeiting, 5%). The best underlying value per category achieves the full weight-ing, while the worst scores zero. All other values are calculated relative to the bestand worst scores. Surveys are published in the March and September issues of thismonthly magazine.

Standard and Poor’s (S&P’s) provides weekly updates on the credit ratings ofsovereign issuers in 77 countries and territories. Sovereign ratings are notcountry ratings as they address the credit risks of national governments, notthe credit risk of other issuers. However, sovereign ratings set the benchmark forthe ratings assigned to other issuers in the country. S&P’s provides short- andlong-term ratings, as well as a qualitative outlook on the sovereign’s domestic andforeign currency reserves. Ratings are provided for seven major areas, namely long-term debt, commercial paper, preferred stock, certificates of deposit, money marketfunds, mutual bond funds, and the claims-paying ability of insurance companies.The determination of credit risk incorporates political risk (the willingness of a

Table 18. Frequency of Risk Component Variables Used in Country Risk Ratings.

Risk components used Frequency

4 103 42 31 0Total 17

Table 19. Agency Data Used.

Agency Frequency

Institutional Investor 13Euromoney 6Moody’s 2Standard and Poor’s 2International Country Risk Guide 2Economist Intelligence Unit 1Political Risk Services 1

Note: Some studies used data from more than one agency.

16 HOTI AND MCALEER


UNCORRECTE

DPROOF

government to service its debt obligations) and economic risk (the government’sability to service its debt obligations) (Howell, 2001). Foreign currency issuerratings are also distinguished from local currency issuer ratings to identify thoseinstances where sovereign risk makes them different for the same issuer. Quantita-tive letter ratings range from C (lowest) to AAA (highest). The rating outlookassesses the potential direction of a long-term credit rating over the intermediate tolonger term. In determining a rating outlook, consideration is given to any changesin the economic and/or fundamental business conditions.

Moody’s provides sovereign credit risk analysis for more than 100 nations,virtually every one of which participates in the world’s capital markets. For eachnation, Moody’s publishes several different types of ratings to capture divergentrisks, including country ratings for both short- and long-term foreign currencysecurities. In establishing country risk, Moody’s analysts assess both political andeconomic variables to derive country risk ratings, which act as sovereign ceilingsor caps on ratings of foreign currency securities of any entity that falls under thepolitical control of a sovereign state (Howell, 2001). Country risk ratings accountfor foreign currency transfer risk and systemic risk in the nation. Using Moody’sAaa to C rating scale, foreign currency long-term government bonds and domes-tic currency long-term government bonds are rated. Local currency guidelineratings, which indicate the highest rating level likely for debt issues denominatedin local currency, are also provided.

Political Risk Services (PRS) provides reports for 100 countries. Each reportassesses potential economic, financial and political risks to business investmentsand trade. Country reports are the only source for risk forecasts and analysisbased on the PRS rating system, which assesses different political scenarios. PRSprovides a political risk model with three industry forecasts at the micro level,namely financial transfers (banking and lending), foreign direct investment (suchas retail, manufacturing, and mining), and exports to the host country market.The 100 reports are revised on a quarterly basis (http://www.prsgroup.com/commonhtml/methods.html).

Economist Intelligence Unit (EIU) publishes country risk reports that areavailable quarterly with monthly updates. These reports summarise the riskratings for all 100 key emerging and highly indebted countries that are monitoredby the Country Risk Service (CRS). The CRS risk rating methodology examinestwo different types of risk: (1) country risk, as determined by (with weights inparentheses) political (22%), economic policy (28%), economic structure (27%),and liquidity (23%) factors; and (2) specific investment risk. Three different typesof specific investment risk are currency risk (associated with accepting foreignexchange exposure against the US dollar), sovereign debt risk (associated withforeign currency loans to sovereign states), and banking sector risk (associatedwith foreign currency loans to banks). These specific investment risk ratings arealso determined by the same four factors, with different weights. For currencyrisk, economic policy is the most heavily weighted factor at 65%, with economicstructure, political, and liquidity factors having weights of 17%, 14%, and 4%,respectively. In the case of sovereign debt risk, liquidity has the highest weight at



UNCORRECTE

DPROOF

31%, with economic policy and economic structure each being weighted at 27%,and the political factor at 15%. Finally, for banking sector risk, economic structureis the most heavily weighted at 44%, with economic policy, liquidity, and politicalfactors weighted at 35%, 15%, and 6%, respectively (http://store.eiu.com).

Table 20 examines the 27 studies concerned with debt rescheduling, in whichthree types of variables were used, namely economic, financial and political. Theeconomic and financial variables were used in each of the 27 studies, whereaspolitical variables were used in only 9 studies. Table 21 presents the number ofvariables used in debt rescheduling, as well as their frequency. All three variableshave been used in 9 studies, two of the three variables were used in the remaining18 studies, and no study used only one of the three variables.

5. Comparison of ICRG Country Risk Ratings

Since January 1984, the International Country Risk Guide (ICRG) has beencompiling economic, financial, political and composite risk ratings for 90 coun-tries on a monthly basis. As of October 2002, the four risk ratings were availablefor a total of 140 countries and 144 entries, the extra four entries relating to theformer sovereign states of Czechoslovakia, East Germany, West Germany andthe USSR. According to the ICRG, its risk ratings have been cited by experts atthe IMF, World Bank, United Nations, and other international institutions, as astandard against which other ratings can be measured. The ICRG has beenacclaimed by publications such as Barron’s and The Wall Street Journal for thestrength of its analysis and rating system.

Several issues relating to the ICRG coverage of the listed countries should beemphasised. Some sovereign states, such as the former Soviet Union republics andthe former Communist Block countries, have been covered only recently. Further-more, structural changes are, in general, not accommodated in the risk ratings.The ICRG rating system was adjusted in late-1997 to reflect the changing inter-national climate created by the ending of the Cold War. Prior to this structuralchange, the financial risk ratings were entirely subjective because of the lack ofreliable statistics. By 1997, the risk assessments were made by the ICRG on thebasis of independently generated data, such as from the IMF, which could bereferenced consistently over time.

Until the dissolution of the former Federal Republic of Yugoslavia, ICRGcovered Yugoslavia which comprised all six republics. After the dissolution,

Table 20. Types of Variables Used in Debt Rescheduling.

Variables Frequency

Economic 27Financial 27Political 9Number of studies 27

18 HOTI AND MCALEER


UNCORRECTE

DPROOF

Yugoslavia refers to the currently constituted Federal Republic of Yugoslavia,comprising the Republic of Montenegro and the Republic of Serbia, whichincludes the UN-administered southern province of Kosovo and the northernprovince of Vojvodina. Since December 1998, ICRG has been covering separatelytwo of the former Yugoslavian republics, namely Croatia and Slovenia, which arenow internationally recognised sovereign states. Data for the other two newsovereign states, namely Bosnia-Herzegovina and the Former Yugoslav Republicof Macedonia, are not currently available. The ICRG coverage of the former Eastand West Germany also merits discussion. After the fall of the Berlin Wall inNovember 1989, East and West Germany were reunited, so there is only oneentry for Germany in the ICRG series from October 1990. Data for the formerWest Germany and East Germany are available separately for January 1984 toSeptember 1990 and June 1984 to September 1990, respectively.

The ICRG rating system comprises 22 variables representing three major com-ponents of country risk, namely economic, financial and political. These variablesessentially represent risk-free measures. There are 5 variables representing each ofthe economic and financial components of risk, while the political component isbased on 12 variables.

Economic risk rating measures a country’s current economic strengths andweaknesses. In general, when a country’s strengths outweigh its weaknesses itpresents a low economic risk, and when its weaknesses outweigh its strengths thecountry presents a high economic risk. This permits an assessment of the ability tofinance its official, commercial, and trade debt obligations. The 5 economicvariables, and the range of risk points assigned to each, are as follows:

(i) GDP per Head of Population (0–5);(ii) Real Annual GDP Growth (0–10);(iii) Annual Inflation Rate (0–10);(iv) Budget Balance as a Percentage of GDP (0–10);(v) Current Account Balance as a Percentage of GDP (0–15).

Financial risk rating is another measure of a country’s ability to service itsfinancial obligations. This rating assesses a country’s financial environment basedon the following 5 financial variables and their associated risk points:

(i) Foreign Debt as a Percentage of GDP (0–10);(ii) Foreign Debt Service as a Percentage of Export in Goods and Services (0–10);

Table 21. Frequency of Types of Variables Used in Debt Rescheduling

Risk components used Frequency

3 92 181 0Total 27



UNCORRECTE

DPROOF

(iii) Current Account as a Percentage of Export in Goods and Services (0–15);(iv) Net Liquidity as Months of Import Cover (0–5);(v) Exchange Rate Stability (0–10).

Political risk rating measures the political stability of a country, which affects thecountry’s ability and willingness to service its financial obligations. The 12 politicalrisk variables, and the range of risk points assigned to each, are as follows:

(i) Government Stability (0–12);(ii) Socio-economic Conditions (0–12);(iii) Investment Profile (0–12);(iv) Internal Conflict (0–12);(v) External Conflict (0–12);(vi) Corruption (0–6);(vii) Military in Politics (0–6);(viii) Religious Tensions (0–6);(ix) Law and Order (0–6);(x) Ethnic Tensions (0–6);(xi) Democratic Accountability (0–6);(xii) Bureaucracy Quality (0–4).

Using each set of variables, a separate risk rating is created for the threecomponents. The 5 variables for the economic risk rating are weighted equallyto give a score of 50 points, the 5 variables for the financial risk rating areweighted equally to give a score of 50 points, and the 12 variables for the politicalrisk rating are weighted equally to give a score of 100 points. As the compositerisk rating is obtained by dividing the sum of the three component risk ratings by2, the economic and financial components account for 25% each and the politicalcomponent accounts for 50% of the composite risk rating.

In all cases, the lower (higher) is a given risk rating, the higher (lower) is theassociated risk. In essence, the country risk rating is a measure of country credit-worthiness. The range of the ICRG risk ratings for economic, financial, political andcomposite risk are 0–50, 0–50, 0–100, and 0–100, respectively. In order to facilitatedirect comparison, in this paper the range of the four risk ratings is given as 0–100.

5.1. Twelve Selected Countries

The risk ratings and volatilities are discussed for twelve representative developingcountries, namely Albania, Argentina, Chile, Cuba, Indonesia, Iraq, Malaysia,Mexico, Romania, Saudi Arabia, South Africa, and Zimbabwe. Following theICRG classification method, Table 22 groups the countries in pairs according totheir geographic regions. The twelve countries represent six geographical regions,namely East Europe (Albania, Romania), South America (Argentina, Chile),North and Central America (Cuba, Mexico), East Asia and the Pacific (Indonesia,Malaysia), Middle East and North Africa (Iraq, Saudi Arabia), and Sub-SaharanAfrica (South Africa, Zimbabwe). Data for these countries have been collectedsince January 1984, apart from Albania and Cuba, for which the data are available

20 HOTI AND MCALEER


UNCORRECTE

DPROOF

from October 1985, and Romania, for which the data are available from August1984. Each of these countries generally has a low risk rating for each of the fourcategories, which is consistent with low creditworthiness and high associated risk.

5.2. Risk Rating Indexes and Volatilities

Risk rating indexes and volatilities for the twelve representative countries are givenin Figures 1a–12a. For each country, the risk rating indexes and volatilities aredenoted ECO-R, FIN-R, POL-R, and COM-R for the economic, financial, poli-tical and composite risk rating indexes, respectively. Defining volatility as thesquared deviation of each observation from the respective sample mean risk ratingindex, the four volatilities are denoted ECO-V, FIN-V, POL-V, and COM-V.

Descriptive statistics for the four risk ratings by country are given in Table 23,in which the twelve countries are ranked according to their means for theeconomic, financial, political and composite risk ratings. In this group of coun-tries, Iraq has the lowest mean risk ratings in all four risk categories, and hence isranked last, while Malaysia has the highest mean risk ratings in all four riskratings, and hence is ranked first. The rankings are generally similar across thefour risk ratings, with a mean range of 3 and a mode of 2. Argentina (3–9),Indonesia (5–11) and Saudi Arabia (2–8) have the highest range of 6 from thelowest to the highest ranking across the four risk ratings. In terms of the meanrank for the four risk ratings, Malaysia is followed by Chile, {Mexico, SaudiArabia}, South Africa, Argentina, {Indonesia, Romania}, Albania, Zimbabwe,Cuba, and Iraq.

The risk rating indexes and associated volatilities for the twelve countries aregiven in Figures 1a–12a. There are substantial changes in the means of the riskrating indexes, as well as in their associated volatilities. Information on theeconomic and political profiles and backgrounds for the twelve representativecountries has been obtained from three sources, namely the US Department ofState: Countries and Regions [http://www.state.gov/countries/], the AustralianDepartment of Foreign Affairs and Trade: Country, Economy and RegionalInformation [http://www.dfat.gov.au/geo/index.html], The Economist: CountryBriefings [http://www.economist.com/countries/], and The World Factbook 2002,prepared by the Central Intelligence Agency [http://www.odci.gov/cia/publica-tions/factbook/index.html].

Table 22. ICRG Classification of Countries by Geographical Region.

Country pairs Selected from geographic region

Albania, Romania East EuropeArgentina, Chile South AmericaCuba, Mexico North and Central AmericaIndonesia, Malaysia East Asia and the PacificIraq, Saudi Arabia Middle East and North AfricaSouth Africa, Zimbabwe Sub-Saharan Africa



UNCORRECTE

DPROOF

Table 23. Descriptive Statistics for Risk Ratings by Country.

Country Risk ratings Mean SD Skewness Minimum Maximum Ranking

Albania Economic 47.43 14.57 �0.64 16 74 10Financial 63.57 6.86 �1.28 42 70 7Political 61.17 5.16 �0.82 46 71 7Composite 58.33 6.52 �1.18 41 69 7

Argentina Economic 53.33 19.45 �0.02 21 84 8Financial 52.23 20.27 �0.41 16 78 9Political 66.35 8.32 �0.25 50 78 3Composite 59.56 13.45 �0.35 36 76 6

Chile Economic 67.46 12.95 �0.56 41 84 4Financial 73.32 12.93 �0.96 45 86 2Political 65.19 12.27 �0.53 43 83 4Composite 67.79 12.05 �0.81 44 84 2

Cuba Economic 44.09 15.75 0.38 24 72 11Financial 48.89 11.42 �0.03 32 64 11Political 59.12 4.44 0.11 52 69 9Composite 52.81 8.20 �0.02 41 65 10

Indonesia Economic 66.59 9.47 �1.96 36 77 5Financial 64.49 16.84 �0.10 36 88 6Political 50.78 8.98 0.43 39 67 11Composite 58.16 9.67 0.07 41 72 8

Iraq Economic 42.33 11.22 �0.51 21 59 12Financial 29.07 17.68 0.59 4 66 12Political 32.54 5.82 �1.28 16 41 12Composite 34.12 7.18 0.35 20 49 12

Malaysia Economic 78.97 5.59 �0.77 61 88 1Financial 76.63 12.72 �0.67 52 90 1Political 69.46 5.80 �0.16 57 82 1Composite 73.63 5.83 �0.36 63 83 1

Mexico Economic 60.97 8.02 0.00 45 80 6Financial 67.48 13.59 �0.43 36 88 4Political 68.03 3.46 �0.27 60 78 2Composite 66.13 6.13 �0.71 52 75 4

Romania Economic 53.51 8.97 �0.24 30 68 7Financial 54.39 13.52 �0.38 30 72 8Political 61.17 9.61 �0.16 45 78 6Composite 57.56 6.52 �0.10 47 70 9

Saudi Arabia Economic 75.33 5.99 �0.53 56 89 2Financial 72.86 15.94 �0.41 46 92 3Political 60.74 7.83 �0.39 45 73 8Composite 67.42 8.36 �0.36 52 81 3

South Africa Economic 68.72 4.19 0.08 59 77 3Financial 66.90 9.75 �0.56 42 82 5Political 64.02 7.74 �0.15 49 77 5Composite 65.91 6.89 �0.32 51 77 5

Zimbabwe Economic 51.11 9.77 �1.22 22 65 9Financial 51.98 6.18 0.59 43 67 10Political 53.80 9.41 �0.08 34 68 10Composite 52.67 6.96 0.02 38 66 11

22 HOTI AND MCALEER


UNCORRECTE

DPROOF

In Figure 1a, the four risk ratings reflect the transition market economy ofAlbania, which ended 44 years of xenophobic communist rule in 1990. Movingfrom a centrally planned to a market system has been difficult due to severeeconomic, social and political inherited problems. Throughout the sample, theeconomic, financial and political indexes followed similar trends, with a discern-able clustering of volatility from 1991 to 2000. The three indexes were low withmild variation until the end of Communist rule. However, changes in the formercommunist bloc by 1990 also affected Albania, with the collapse in social andeconomic life. In 1991, clashes between Communists and their opponents led tothe fall of the regime, with a sharp fall in the three indexes and associatedvolatility peaks. The 1992 Democratic Party government, led by President SaliBerisha, launched an ambitious program which resulted in price and exchangerate liberalisation, fiscal consolidation, monetary restraint, privatisation, enter-prise and financial reform, and high growth. However, progress was stalled in1997 when several pyramid financial schemes collapsed and the indexes dropped.The collapse caused panic and led to the fall of the government, with the SocialistParty coming to power in June 1997. After 1998, the economic and financialindexes rose and remained flat, as the economy recovered from foreign remit-tances, expansion of the construction and service industries, and an increase inseaside tourism. Major unresolved problems include soaring unemployment, highinflation, dilapidated infrastructure, unexploited natural resources, low foreigninvestment, high trade deficits, and inefficient energy production. In an effort to

0

200

400

600

800

10000

20

40

60

80

86 88 90 92 94 96 98 00

ECO-R ECO-V

0

100

200

300

400

500 40

50

60

70

80

86 88 90 92 94 96 98 00

FIN-R FIN-V

04080

120160200240 45

505560657075

86 88 90 92 94 96 98 00

POL-R POL-V

0

100

200

300

40040

50

60

70

86 88 90 92 94 96 98 00

COM-R COM-V

ECO-V ECO-R FIN-V FIN-R

COM-VPOL-RPOL-V COM-R

Figure 1a. Risk Rating Indexes and Volatilities for Albania.

Note: Economic (ECO), Financial (FIN), Political (POL) and Composite (COM) risk rating indexesand their associated volatilities are denoted by R and V, respectively.



UNCORRECTE

DPROOF

promote regional trade, in 2001 Albania agreed with Macedonia, Bulgaria,Croatia, Yugoslavia, and Romania to establish a free trade area in south centralEurope by 2004. Political instability was high from 1997–2000, after which thepolitical index was consistently flat. While democratic reforms continue, govern-ment and judicial corruption, and widespread organised crime, remainunchecked. Albania pursues greater Euro-Atlantic integration and restored itsrelations with Yugoslavia following the ouster of Milosevic, but the status ofKosovo remains a key unresolved issue. As an overall measure of country risk,the composite index reflects the trends and volatility in the three component riskindexes.

The four risk rating indexes for Argentina are given in Figure 2a. Argentina isrich in resources and has a well-educated workforce, but economic growth hasgenerally not matched expectations. From 1880 to the 1930s, Argentina was oneof the world’s ten wealthiest countries owing to the rapid expansion of agricultureand foreign investment in infrastructure. However, over the past 25 years,Argentina has struggled with military dictatorship, the war over the Falkland Islands,and severe economic difficulties. There is a similar pattern, with a discernableclustering of volatility, in the economic and political indexes, starting at very lowvalues and following a generally increasing trend until 1999, after which they havereturned to their original values. Similarly, the financial index increased to 1995and then decreased, with an associated clustering of volatility. The low indexes inthe 1980s were the result of protectionist and populist economic policies in the

0200400600800

10001200

20

40

60

80

100

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

400

800

1200

1600

0

20

40

60

80

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

04080

120160200240280

40

50

60

70

80

84 86 88 90 92 94 96 98 00

POL-R POL-V

0100200300400500600

30

40

50

60

70

80

84 86 88 90 92 94 96 98 00

COM-R COM-V


COM-RCOM-VPOL-RPOL-V

Figure 2a. Risk Rating Indexes and Volatilities for Argentina.


24 HOTI AND MCALEER


UNCORRECTE

DPROOF

post-war era that led to economic stagnation and hyperinflation. When CarlosMenem was elected President in 1989, he abandoned the former policies in favourof market economics and liberalisation, resulting in a period of rapid growth. Hisfailure to sustain the fiscal and structural reforms in his second term from 1995 to1999 left the economy vulnerable to the 1994 Mexican ‘Tequila’ crisis, the 1997 SouthEast Asian crisis, the Russian default of 1998, and the Brazilian devaluation of 1999.These shocks led to a higher cost of foreign borrowing and less competitive exports.An IMF bailout package of nearly $40 billion in late 2000, involving tax rises and cutsin social welfare programs, resulted in a political crisis that caused the government tocollapse amid violent protests. A new government was elected in January 2002, whenArgentina abandoned the quasi-currency board system, which pegged the peso atparity to the US dollar for over 10 years. The subsequently floated peso increased asense of expropriation for depositors, so that almost all dollar loans were converted tothe peso at parity. Consequently, bank balance sheets and reputations were destroyed,and the number of banks and the scale of banking operations shrank significantly. Thebanking system, formerly one of the strongest in Latin America, has been decimated.Overall, the composite risk index closely reflects the trends and volatility in theeconomic and political risk indexes.

Figure 3a presents the risk rating indexes for Chile, one of the world’s mostopen economies. Reforms such as privatisation, liberalisation and deregulation oftrade and investment, were initiated by the military government and continued bysubsequent democratic administrations. The economic, financial, and politicalindexes were low and flat until 1987, after which they followed an increasingtrend and then decreased, with virtually no change in the financial index between1991 and 1997. There are discernable clusterings of volatility for the three com-ponent risk indexes. The economic index fell in 1998 and remained low in 1999 asa result of the recession due to the global downturn. However, it increased aseconomic recovery began in 2000, but followed a slight declining trend to 2002,while both domestic and foreign investments fell, unemployment rose, and eco-nomic growth slowed. In early 2002, the government committed to undertakemicroeconomic reforms to create new incentives for private investment. After aperiod of stability, the financial index also fell in 1997 and varied around the mid70s until 2002. The government implemented further liberalisation of capitalmarkets in 2001, leading to a rise in the index until the end of the sample, andwas repaying foreign debt by 2002, which was low by Latin American standards.Changes in the political scene in December 1988, when Augusto Pinochet failed towin a referendum, led to democratic elections in December 1989. There have sincebeen three consecutive presidents from the Concertacion de Partidos por laDemocracia coalition. After the defeat of Pinochet’s military regime, the consti-tution was amended to ease provisions for future amendments and diminish therole of the National Security Council by equalising the number of civilian andmilitary members. However, further constitutional reforms are still necessary tocomplete the full transition to democracy. Since the return to democracy, Chilehas become an active participant in the international political arena, but thepolitical index decreased in 1998 before increasing in 2000 due to the detention



UNCORRECTE

DPROOF

of Pinochet in the UK in response to an extradition request by Spain. This periodwitnessed demonstrations by supporters and opponents of Pinochet, which led toclashes with the police. Overall, the composite risk index reflects the trends andvolatility in the three component risk ratings.

The risk rating indexes for Cuba are presented in Figure 4a. Cuba has been acommunist country since Fidel Castro led his army to victory in 1959. During theCold War, Cuba relied on strong Soviet support, and built reputable health andeducation systems. Amid the USA trade sanctions, Castro failed to diversifythe economy, which continues to depend on sugar exports. Economic hardshipwas heightened by the high price of foreign financing. Cuba relies heavily onshort-term loans to finance imports. The government defaulted on most of itsinternational debt in 1986 and does not have access to credit from internationalfinancial institutions, such as the World Bank. There was a falling trend in theeconomic, financial and political indexes until 1992, after which the economic andfinancial indexes rose, but the political index increased and fell before increasingagain in the last two years. The falling economic and financial indexes were due tothe withdrawal of aid from the former Soviet Union, as well as domestic incom-petence, which led to a severe economic recession in 1990–1992. In response to theeconomic crisis, in 1993 and 1994 the government launched some market reforms,including opening tourism, allowing foreign investment from Canada, Europe,and Latin America, legalising the dollar, and authorising self-employment for 150occupations, which resulted in modest economic growth. However, living

0

200

400

600

80040

50

60

70

80

90

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

200

400

600

800

1000

40

50

60

70

80

90

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

100

200

300

400

500 40

50

60

70

80

90

84 86 88 90 92 94 96 98 00

POL-R POL-V

0100200300400500600 40

50

60

70

80

90

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 3a. Risk Rating Indexes and Volatilities for Chile.


26 HOTI AND MCALEER


UNCORRECTE

DPROOF

standards by 2000 have remained well below 1989 levels, as lower sugar andnickel prices, higher petroleum costs, a post-September 11, 2001 decline in tourism,and a devastating November 2001 hurricane, created new economic pressures.Furthermore, the legalisation of the US dollar created a serious economic gapbetween those with and without access to dollars. Continuing hardships led to anincrease in prostitution, corruption, black market activity, and desperate effortsto escape the country. After the Cold War, Cuba abandoned monetary supportfor guerrilla movements in Latin America and Africa, but maintained relationswith several guerrilla and terrorist groups. The USA Helms-Burton legislationagainst trade with, and investment in, Cuba in 1996 led to a decreasing politicalindex. There is a discernable clustering of volatility in the three component riskindexes, with the composite risk index reflecting the trends and volatility in theeconomic and financial risk indexes.

Figure 5a presents the four risk ratings for Indonesia, which has a market-based economy with significant government participation and ownership.Indonesia has experienced unprecedented turmoil since 1997 due to the South-EastAsian financial crisis, the fall of President Suharto after 32 years, the first freeelections since the 1960s, the loss of East Timor, independence demands from restiveprovinces, bloody inter-ethnic and religious conflicts, and unending corruptionscandals. Political and judicial reforms and the restructuring of the banking sectorand offshore debt are crucial steps towards economic recovery and growth. Theeconomic index had a slightly increasing trend until 1997, when it fell due to the

0

200

400

600

800

20

30

40

50

60

70

80

86 88 90 92 94 96 98 00

ECO-R ECO-V

0

100

200

300 30

40

50

60

70

86 88 90 92 94 96 98 00

FIN-R FIN-V

0

20

40

60

80

100

50

55

60

65

70

86 88 90 92 94 96 98 00

POL-R POL-V

0

40

80

120

160 40455055606570

86 88 90 92 94 96 98 00

COM-R COM-V



Figure 4a. Risk Rating Indexes and Volatilities for Cuba.




UNCORRECTE

DPROOF

severe financial crisis, varied around the 40s until 2000, when it increased by almost30 points, and was flat at 70 until 2002, with discernable volatility from 1996 to 2000.Prior to 1997, the government removedmost regulatory obstacles to the external andfinancial sectors, which led to increased employment, higher growth in the non-oilexport sector, and a growth rate of 7% from 1987 to 1997. While economic recoveryafter the crisis has been slow, consumer confidence and exports increased and therupiah was stabilised. Indonesia had a low financial index until 1988, after which itincreased substantially, remained in the high 80s until 1992, followed a downwardtrend until the 1997 crisis, fell by almost 40 points until 1998, increased in 1999, andremained highly volatile until 2002. While the rupiah has stabilised since 2001, theinvestment climate has remained troubled due to the lack of legal protection,confusion over regional autonomy policies and fiscal decentralisation, uneven imple-mentation of economic reforms, and tax and labour issues. The political scene hasalso been volatile, with the index improving substantially from 1988 to 1997, afterwhich it fell and remained low, but with high variation. Such a fall in the index wasdue to Soeharto, who presided over 32 years of authoritarian rule, having beenforced out in May 1998, amid deepening economic, financial and social crises. In2001, there was a peak in volatility when President Wahid was impeached due tocompetence and replaced by President Soekarnoputri. As a weighted average of thethree component risk indexes, the composite index reflects the trends and volatilityin the financial and political risk indexes.

0

200

400

600

800

100030

40

50

60

70

80

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

200

400

600

800

1000

30

40

50

60

70

80

90

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

40

80

120

160

200

240

280

30

40

50

60

70

84 86 88 90 92 94 96 98 00

POL-R POL-V

0

100

200

300 40

50

60

70

80

84 86 88 90 92 94 96 98 00

COM-R COM-V


POL-V POL-R COM-V COM-R

Figure 5a. Risk Rating Indexes and Volatilities for Indonesia.


28 HOTI AND MCALEER


UNCORRECTE

DPROOF

The four risk rating indexes for Iraq in Figure 6a are all very low. Although Iraqis an oil-rich country, the war with Iran from 1980–1988 seriously damaged its oilexport facilities, depleted foreign exchange reserves, devastated the economy, andleft significant foreign debt. Consequently, the economic index was in the mid 20sin 1987, with discernable volatility prior to mid 1987, but followed a rising trendfrom 1988–1991 as new pipelines were constructed, damaged facilities wererepaired, and oil exports gradually increased. Iraq’s seizure of Kuwait in 1990and the 1991 war with the US-led UN coalition resulted in international sanctions,which drastically reduced economic activity. During this period, the index droppedby 30 points, after which it followed an upward trend, with discernable volatilityafter 1994. A UN oil-for-food program launched in 1996 improved living condi-tions, and Iraq was authorised in 1991 to export unlimited oil quantities to financehumanitarian needs, including food, medicine, and infrastructure. There was notrend and little variation in the financial index to 1990, being very low during thisperiod, with a clustering of volatility in 1990–1991. In the 1980s, financial difficul-ties caused by massive expenditures on the war with Iran led to the implementationof austerity measures, heavy borrowing, and debt rescheduling. From 1991 to 1994,the financial index was virtually flat at 10, after which it had a rising trendassociated with increasing volatility. The end of the war with Iran in 1988 saw arise in the political index, followed by a drop during the Gulf Crisis from 1990 to1992, a clustering of volatility, and then an increasing trend, with little variation.After the Gulf Crisis, the major issue for Saddam Hussein’s regime was to retain

0

100

200

300

400

500 20

30

40

50

60

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

400

800

1200

1600

0

20

40

60

80

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

100

200

300 15202530354045

84 86 88 90 92 94 96 98 00

POL-R POL-V

0

50

100

150

200

250

10

20

30

40

50

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 6a. Risk Rating Indexes and Volatilities for Iraq.




UNCORRECTE

DPROOF

power while overcoming international isolation, and dealing with calls to surrenderweapons of mass destruction and submit to UN inspections. The regime refused tocooperate fully with the UN inspectors, and allowed no inspections in Iraq after1998. Relations with the Arab nations have been unstable, and Iraq has embracedthe most extreme anti-Israeli position. In 2002, US President George W. Bushdeclared Iraq to be part of ‘an axis of evil’, which led to rising tensions in globalpolitics. Overall, the composite risk index has followed a similar trend to that of thepolitical index, but reflects the volatility in all three component risk ratings.

Figure 7a presents the four risk ratings for Malaysia. After independence in1957, the economy was based on two commodities, rubber and tin, but theeconomic performance was impressive for the following 40 years. New foreignand domestic investments from the early 1980s to mid 1990s played a significantrole in transforming the economy from a commodity-based to a manufacturingsystem. Although one of the world’s largest exporters of semiconductors andtargeted toward being a leading producer and developer of high-tech products,the sustained high growth ended with the Asian crisis in 1997. The economicindex followed an increasing trend until the economic and financial crises in 1997,after which it fell, reaching the low 60s in 1998, with an associated peak involatility. During 1998, the government focused on expansionary measures todeal with the crises. A range of capital controls was implemented to restrict theflow of capital in and out of Malaysia, and the ringgit was pegged against the USdollar. The economic index increased from 1999 to 2000 as capital controls were

0

100

200

300

40060

70

80

90

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

200

400

600

800

50

60

70

80

90

100

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

40

80

120

160 55606570758085

84 86 88 90 92 94 96 98 00

POL-R POL-V

0

40

80

120

160

60

65

70

75

80

85

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 7a. Risk Rating Indexes and Volatilities for Malaysia.


30 HOTI AND MCALEER


UNCORRECTE

DPROOF

eased to restore foreign investor confidence and the economy recovered. A down-ward trend in the index after 2000 reflected the impact of the global downturn onexports and economic growth. The financial index was also affected by the crisesin 1997. However, the index started to increase in the same year, with anassociated peak in volatility, and remained flat after 1999. Malaysia is a multi-ethnic federation, with the Malay community benefiting from positive discrimin-ation in business, education and the civil service. However, the ethnic Chinese holdeconomic power and are the wealthiest community. A decreasing trend in thepolitical index to 1988 was followed by an increasing trend until the 1997 Asiancrisis, after which the index fell and followed a slight upward trend from 1998,with discernable volatility throughout the sample. A serious challenge remains tosustain political stability amid the economic downturn and the ethnic wealth gap.As a founding member of ASEAN, Malaysia views regional cooperation asthe basis for its foreign policy. Overall, the composite risk index closely reflectsthe trends in the financial and political risk indexes, and is less volatile than thepolitical index.

Figure 8a gives the four risk ratings for Mexico, which two decades ago wasclosed to foreign investment and trade, with strong government participation.Now one of the world’s most trade-dependent countries, Mexico has Free TradeAgreements with the USA, Canada, EU, and others. There is a large oil sector,which provides a third of government revenues, but is not enough for economic

0

100

200

300

400

40

50

60

70

80

90

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

200

400

600

800

1000 30405060708090

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

20

40

60

80

100

55

60

65

70

75

80

84 86 88 90 92 94 96 98 00

POL-R POL-V

0

50

100

150

200

250

50556065707580

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 8a. Risk Rating Indexes and Volatilities for Mexico.




UNCORRECTE

DPROOF

prosperity. However, the country is undergoing substantial change, as the 1997elections resulted in a victory for the combined opposition, breaking the one-party system with a democratic facade. The 2000 presidential elections confirmedthe development, as an opposition candidate, Vicente Fox, became president forthe first time. Massive external debt default in 1982, the 1984 oil price crisis, andaccession to GATT in 1986, are reflected in movements in the economic, financialand political indexes, which followed a declining trend to 1986. The upward trendafter 1986 was due to economic reforms by the government, including trade andinvestment liberalisation, privatisation, deregulation and fiscal consolidation.From 1988 to 1994, President Salinas began a process of restructuring theeconomy, which was continued by the Zedillo administration from 1994 to2000. These reforms and growing ties with the USA led to a period of relativelystrong growth and stability in the economy. The 1994–1995 peso crisis led to a fallin the financial index, which rose and fell in 1997. However, the economyrecorded a contraction in 2001, which affected the economic and political indexes,but not the financial index. While the Argentine debt crisis had no significanteffect on Mexico, the downturn was attributed to economic factors in the USAand the events of September 11, 2001 that led to caution towards US bordertrade, in particular, and a significant reduction in tourism. The socio-politicalconflict in the southern province of Chiapas remains unresolved. In response topressures for greater rights for indigenous people, President Fox has shown awillingness to deal with the demands of guerrillas. Traditionally, Mexico’s foreignpolicy has been based on non-intervention and self-determination. Overall, thecomposite risk index reflects the trends and volatility in the economic, financialand political risk indexes.

The four risk rating indexes in Figure 9a reflect Romania, with rich agriculturallands, diverse energy sources, large manufacturing industrial base, an educated andwell trained labour force, and great potential for tourism in the Black Sea and themountains. Romania has moved slowly from its communist past relative to itsEastern European neighbours. Despite the 1989 uprising that ended the severe ruleof President Nicolae Ceausescu, the former communists remain a dominant force innational politics. Since the change in regime, successive governments have tried tobuild a Western style market economy. The pace of restructuring has been slow dueto a persistent failure to undertake structural reforms, with periods of high growthfollowed by high inflation and economic imbalance. Further fiscal consolidation,restructuring and privatisation of large state enterprises and utilities remain majoreconomic policy issues. The economic index had a declining trend to 1991, afterwhich it followed a generally increasing trend, with greater volatility. As reflected inthe falling trend, Romania entered a period of deep recession in mid-1997, duemainly to industrial restructuring. Economic recovery started in 1999 and, thoughinflation remained high, foreign investment increased slowly and trade with theWest also grew. The financial and political indexes decreased until the fall of theregime in the late 1980s and followed increasing trends until 1997, with noticeablevolatility, especially for the political index. There was a fall in the financial index in1997 as a result of the recession, after which it was unstable until early 2000, with a

32 HOTI AND MCALEER


UNCORRECTE

DPROOF

noticeable volatility clustering from mid 1997 to early 2001 owing to foreignexchange shortages and a poorly developed financial sector. Two major politicalchanges occurred in Romania after 1989 which led to downward movements in thepolitical index and increasing volatility, namely the electoral defeat in 1996 of theformer Communists, who came to power after Ceausescu’s demise, and the 2000elections, which saw the former Communists returned to power. Under new Pre-sident Illiescu, the political situation has stabilised, as shown by the flat politicalindex, and low associated volatility. Romania aims to strengthen relations with theWest, particularly the USA and EU. Generally, the composite risk index closelyreflects the trends and volatility in the financial and political indexes.

Figure 10a gives the four risk rating indexes for Saudi Arabia. Ruled as anabsolute monarchy since its creation in 1932, Saudi Arabia emerged from anunderdeveloped desert kingdom to be one of the wealthiest Middle Easternnations due to its vast oil resources. The oil reserves are the world’s largest andaccount for more that 90% of national exports and almost 75% of governmentrevenues. Since 1970, economic development has been based on five-year plansfocusing on infrastructure, education, health and social services, private enter-prise, foreign investment, and consolidation of national defence. The 2000–2004plan emphasised economic diversification, a greater role for the private sector,and expansion of the oil and gas industry. There are noticeable structural changesin the financial and political indexes, but not the economic index, which followed

0

100

200

300

400

500

600

20

30

40

50

60

70

86 88 90 92 94 96 98 00

ECO-R ECO-V

0

100

200

300

400

500

60020

30

40

50

60

70

80

86 88 90 92 94 96 98 00

FIN-R FIN-V

0

100

200

300

40

50

60

70

80

86 88 90 92 94 96 98 00

POL-R POL-V

0

50

100

150

45

50

55

60

65

70

86 88 90 92 94 96 98 00

COM-R COM-V



Figure 9a. Risk Rating Indexes and Volatilities for Romania.




UNCORRECTE

DPROOF

a generally increasing trend, with discernable volatility throughout the sample.From 1997 to 1999, low oil prices slowed state-led industrial development, caus-ing the index to fall. In 2000, the index started to increase, due to high oil prices,but fell again in 2001, due to the sharp fall in oil prices. Recently, the governmentfocus has been on providing sustainable utilities, services and jobs for a growingpopulation. The Gulf Crisis in early 1991 was associated with structural changesin the financial and political indexes. Such changes indicate that, since 1991,Saudi Arabia has been safer with respect to financial and political risk. Politicaland judicial institutions are ruled by Islamic principles, with the King holdingabsolute legislative and executive power. The government faces a number ofpolitical challenges, with an overstaffed civil service, conservative educationsystem, and widespread corruption and waste. There are fears that the largeunemployed youth could be drawn to radical Islamic groups. Saudi Arabia’sforeign policy objectives are to maintain security and a dominant position inthe Arabian Peninsula, defend general Arab and Islamic interests, promote soli-darity among Islamic governments, and maintain cooperative relations withmajor oil-producing and oil-consuming nations. The government has acted as amediator in regional crises and in Israeli-Palestinian peace negotiations. As anoverall measure of country risk, the composite index reflects the trends andvolatility in the financial and political indexes.

The four risk rating indexes in Figure 11a are for South Africa, a middle-incomedeveloping country with daunting economic problems inherited from the apartheid

0

100

200

300

40050

60

70

80

90

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

200

400

600

800405060708090100

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

50

100

150

200

250 40

50

60

70

80

84 86 88 90 92 94 96 98 00

POL-R POL-V

04080

120160200240 50

60

70

80

90

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 10a. Risk Rating Indexes and Volatilities for Saudi Arabia.


34 HOTI AND MCALEER


UNCORRECTE

DPROOF

era, especially poverty and little economic power among disadvantaged groups.South Africa is productive and industrialised, with a high division of labourbetween the formal and informal sectors, and an uneven wealth and incomedistribution. A diverse manufacturing industry has made it a world leader in severalspecialised sectors, including railway rolling stock, synthetic fuels, and miningequipment and machinery. The Growth, Employment and Redistribution(GEAR) strategy was directed toward open markets, privatisation and a favourableinvestment climate, but had mixed results. However, great progress has been madein restructuring the economic system, which was based on import substitution, hightariffs and subsidies, anti-competitive behaviour, and government intervention. Thedeclining trend in the economic and financial indexes in 1994–1999 was due to lowand unstable growth, high unemployment, skyrocketing crime, corruption, andHIV/AIDS. After 1999, the indexes started to increase as President Mbeki vowedto promote economic growth and foreign investment, and to reduce poverty byrelaxing restrictive labour laws, increasing the pace of privatisation, and reducinggovernment spending. The economy slowed in 2001 as a result of the internationaldownturn. South Africa has a sophisticated financial structure, with a large andactive stock exchange and mild exchange controls. The political index fell to 1987,followed a generally increasing trend to 1997 and a declining trend thereafter, withdiscernable volatility throughout the sample. Ruled by a white minority until 1994,South African activists struggled for much of the last century before succeeding in

0

20

40

60

80

10055

60

65

70

75

80

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0

200

400

600

800

40

50

60

70

80

90

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

0

50

100

150

200

250

40

50

60

70

80

84 86 88 90 92 94 96 98 00

POL-R POL-V

04080

120160200240 50

556065707580

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 11a. Risk Rating Indexes and Volatilities for South Africa.




UNCORRECTE

DPROOF

overthrowing apartheid and extending democracy. After 1994, President Mandela’sleadership encouraged democratic reforms and reconciliation, amid painful legaciesof lawlessness, social disruption and lost education. Under the 1999 government ofPresident Mbeki, economic transformation became a priority. Having emergedfrom the isolation of the apartheid era, South Africa has become an active playerin international politics. Overall, the composite risk index closely reflects the trendsand volatility in the three component risk ratings.

Figure 12a gives the four risk rating indexes for Zimbabwe. Robert Mugabe,the first prime minister and president since 1987, has been the sole ruler anddominated the political landscape since independence, but now presides overchaos, a land crisis and faltering economy. Under proper economic management,the wide range of natural resources should be able to support sustained economicgrowth. The country has large reserves of metallurgical-grade chromite and othercommercial mineral deposits, and has long been the world’s third largest tobaccoexporter. There was a generally declining trend in the economic and politicalindexes, with a noticeable clustering of volatility, while the financial index fol-lowed no trend but was highly volatile. Earlier moves to develop a market-oriented economy led to a reduction in the economic index, with an associatedvolatility peak in 1992. The index followed an increasing trend after 1992, butstarted to decline in 1997, with increasing volatility. Similarly, while the financialindex varied substantially throughout the sample, its associated volatility washigher in the second half of the sample because IMF support was suspended due

0

200

400

600

800

100020

30

40

50

60

70

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

ECO-RECO-V

0

50

100

150

200

250

40

45

50

55

60

65

70

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

FIN-RFIN-V

0

100

200

300

400 30

40

50

60

70

84 86 88 90 92 94 96 98 00

POL-R POL-V

POL-RPOL-V

0

50

100

150

200

250

30

40

50

60

70

84 86 88 90 92 94 96 98 00

COM-R COM-V

COM-RCOM-V

Figure 12a. Risk Rating Indexes and Volatilities for Zimbabwe.


36 HOTI AND MCALEER


UNCORRECTE

DPROOF

to a failure to meet budget targets. Moreover, the economy had been steadilyweakened by excessive government deficits, AIDS, rampant inflation, andextreme income inequality. The government’s land reform program, characterisedby chaos and violence, derailed the commercial sector, which had been a tradi-tional source of exports, foreign exchange and employment. Politically,Zimbabwe had been improving, as shown by the increasing trend in the politicalindex to 1998. However, involvement in the war in the Democratic Republic ofthe Congo, which began in 1998, contributed to domestic woes and caused theindex to enter a steep declining trend, reaching the mid 30s by 2002. In early 1999,Zimbabwe experienced considerable political and economic upheaval, and grow-ing opposition against Mugabe’s regime. Local and international human rightsmonitors have noted a marked increase in human rights abuses since early 2000.Parliamentary elections in mid 2000 and presidential elections in early 2002 wereassociated with violent intimidation against opposition supporters, the pressand judiciary. Overall, the composite risk index closely reflects the trends andvolatility in the economic and political indexes.

5.3. Risk Rating Returns and Volatilities

Risk returns are defined as the monthly percentage change in the respective riskrating indexes. The descriptive statistics for risk returns by country are given inTable 24, and the correlation coefficients for risk returns by country are given inTable 25. For each country the risk returns in Figures 1b–12b are denoted ECO-R,FIN-R, POL-R and COM-R for the economic, financial, political and compositerisk returns, respectively. Defining volatility as the squared deviation of each obser-vation from the respective sample mean risk return, the four volatilities associatedwith the risk returns are denoted ECO-V, FIN-V, POL-V and COM-V, respectively.

Table 24 reports the descriptive statistics for the four risk returns by country.The means of all four risk returns for the twelve countries are close to zerowith standard deviations ranging from 0.0205 (Indonesia) to 0.1117 (Iraq) foreconomic risk returns, 0.0202 (Chile) to 0.1391 (Iraq) for financial riskreturns, 0.0130 (Malaysia) to 0.0558 (Iraq) for political risk returns, and 0.0103(Indonesia) to 0.0486 (Iraq) for composite risk returns. Of the twelve countries,Iraq has the highest standard deviation for the four risk returns. There is nogeneral pattern of skewness for the four risk returns for the twelve countries, withall four risk returns being negatively skewed for Albania and Malaysia, and allpositively skewed for Saudi Arabia. While both the financial and political riskreturns are positively skewed for Iraq and Zimbabwe, only the political risk returnsare positively skewed for Argentina and Romania. Economic risk returns are theonly positively skewed risk returns for Indonesia, but the only negatively skewedrisk returns for South Africa. For Mexico, the financial and composite risk returnsare both negatively skewed, for Cuba only the financial risk returns are negativelyskewed, and only the composite risk returns are negatively skewed for Chile.

Table 25 reports the correlation coefficients for the four risk returns by coun-try. The economic, financial and political risk returns seem to be highly correlated



UNCORRECTE

DPROOF.000

.004

.008

.012

.016

.020–.15

–.10

–.05

.00

.05

.10

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.000

.005

.010

.015

.020

.025

.030

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.00

.01

.02

.03

–.10

–.05

.00

.05

.10

.15

.20

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.002

.004

.006

.008

–.04

.00

.04

.08

.12

84 86 88 90 92 94 96 98 00

COM-R COM-V

FIN-RFIN-VECO-RECO-V


Figure 1b. Risk Returns and Volatilities for Albania.

Note: Risk returns (R) and their associated volatilities (V) refer to the rates of change in the respectiverisk rating indexes.

.00

.02

.04

.06

.08

–.3–.2–.1.0.1.2.3

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

ECO-RECO-V

.00

.02

.04

.06

.08

–.2

–.1

.0

.1

.2

.3

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

FIN-RFIN-V

.00

.01

.02

.03

.04

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

POL-R POL-V

POL-RPOL-V

.000

.002

.004

.006

.008

–.10

–.05

.00

.05

.10

84 86 88 90 92 94 96 98 00

COM-R COM-V

COM-RCOM-V

Figure 2b. Risk Returns and Volatilities for Argentina.


38 HOTI AND MCALEER


UNCORRECTE

DPROOF.00

.01

.02

.03

.04

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.000

.001

.002

.003

.004

.005

.006 –.08

–.04

.00

.04

.08

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.000

.001

.002

.003

.004

.005

–.08

–.04

.00

.04

.08

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.001

.002

.003

.004

.005

–.08

–.04

.00

.04

.08

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 3b. Risk Returns and Volatilities for Chile.


.00

.01

.02

.03

.04

.05

–.3

–.2

–.1

.0

.1

.2

.3

86 88 90 92 94 96 98 00

ECO-R ECO-V

.00

.04

.08

.12–.4

–.2

.0

.2

.4

86 88 90 92 94 96 98 00

FIN-R FIN-V

.000

.002

.004

.006

.008

.010

–.04

.00

.04

.08

.12

86 88 90 92 94 96 98 00

POL-R POL-V

.000

.004

.008

.012

.016

.020

–.10

–.05

.00

.05

.10

.15

86 88 90 92 94 96 98 00

COM-R COM-V



Figure 4b. Risk Returns and Volatilities for Cuba.




UNCORRECTE

DPROOF.00

.05

.10

.15

.20

.25

–.4

–.2

.0

.2

.4

.6

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.04

.08

.12

–.4

–.2

.0

.2

.4

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.00

.01

.02

.03

.04–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.005

.010

.015

.020

.025

.030

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 5b. Risk Returns and Volatilities for Indonesia.


.0

.1

.2

.3

.4

.5

.6–.8

–.4

.0

.4

.8

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

0.0

0.4

0.8

1.2

1.6

–1.0

–0.5

0.0

0.5

1.0

1.5

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.00

.04

.08

.12

.16

–.4

–.2

.0

.2

.4

84 86 88 90 92 94 96 98 00

POL-R POL-V

.00

.02

.04

.06

.08–.3–.2–.1.0.1.2.3

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 6b. Risk Returns and Volatilities for Iraq.


40 HOTI AND MCALEER


UNCORRECTE

DPROOF.000

.005

.010

.015

.020

.025

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.00

.02

.04

.06

.08–.3

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.0000

.0004

.0008

.0012

.0016

.0020

.0024

.0028

–.08

–.04

.00

.04

.08

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.002

.004

.006

.008

–.12

–.08

–.04

.00

.04

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 7b. Risk Returns and Volatilities for Malaysia.


.00

.01

.02

.03

.04

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.00

.01

.02

.03

.04

.05

–.3

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.000

.002

.004

.006

.008

.010

–.08

–.04

.00

.04

.08

.12

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.002

.004

.006

.008

.010

–.12

–.08

–.04

.00

.04

.08

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 8b. Risk Returns and Volatilities for Mexico.




UNCORRECTE

DPROOF.00

.05

.10

.15 –.4

–.2

.0

.2

.4

86 88 90 92 94 96 98 00

ECO-R ECO-V

.0

.1

.2

.3

.4

.5

–.8

–.4

.0

.4

.8

86 88 90 92 94 96 98 00

FIN-R FIN-V

.000

.002

.004

.006

.008

.010

–.08

–.04

.00

.04

.08

.12

86 88 90 92 94 96 98 00

POL-R POL-V

.000

.004

.008

.012

.016

.020

–.15

–.10

–.05

.00

.05

.10

86 88 90 92 94 96 98 00

COM-R COM-V



Figure 9b. Risk Returns and Volatilities for Romania.


.00

.01

.02

.03

.04

.05–.2

–.1

.0

.1

.2

.3

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.00

.01

.02

.03

.04

.05

.06

–.3

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.00

.01

.02

.03

.04

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.004

.008

.012

.016

.020

–.10

–.05

.00

.05

.10

.15

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 10b. Risk Returns and Volatilities for Saudi Arabia.


42 HOTI AND MCALEER


UNCORRECTE

DPROOF.000

.004

.008

.012

.016

.020–.15

–.10

–.05

.00

.05

.10

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.000

.005

.010

.015

.020

.025

.030

–.2

–.1

.0

.1

.2

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.00

.01

.02

.03

–.10

–.05

.00

.05

.10

.15

.20

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.002

.004

.006

.008

–.04

.00

.04

.08

.12

84 86 88 90 92 94 96 98 00

COM-R COM-V

FIN-RFIN-VECO-RECO-V


Figure 11b. Risk Returns and Volatilities for South Africa.


.00

.02

.04

.06

.08–.3–.2–.1.0.1.2.3

84 86 88 90 92 94 96 98 00

ECO-R ECO-V

.00

.05

.10

.15

.20

.25

.30

–.6

–.4

–.2

.0

.2

84 86 88 90 92 94 96 98 00

FIN-R FIN-V

.000

.002

.004

.006

.008

–.12

–.08

–.04

.00

.04

.08

84 86 88 90 92 94 96 98 00

POL-R POL-V

.000

.004

.008

.012

.016

.020–.15

–.10

–.05

.00

.05

.10

84 86 88 90 92 94 96 98 00

COM-R COM-V



Figure 12b. Risk Returns and Volatilities for Zimbabwe.




UNCORRECTE

DPROOF

Table 24. Descriptive Statistics for Risk Returns by Country.

Country Risk returns Mean SD Skewness

Albania Economic 0.0016 0.0853 �2.3776Financial �0.0005 0.0400 �0.8775Political �0.0001 0.0301 �0.5235Composite 0.0002 0.0276 �2.1586

Argentina Economic 0.0026 0.0636 �0.5162Financial 0.0006 0.0585 �3.8034Political 0.0008 0.0207 0.2672Composite 0.0012 0.0222 �1.4046

Chile Economic 0.0021 0.0390 0.3235Financial 0.0021 0.0202 0.2447Political 0.0021 0.0157 0.7824Composite 0.0021 0.0148 �0.2724

Cuba Economic 0.0020 0.0393 0.5590Financial 0.0002 0.0410 �0.8419Political 0.0001 0.0133 1.7161Composite 0.0006 0.0169 2.4361

Indonesia Economic 0.0000 0.0205 2.6154Financial �0.0011 0.0310 �3.3830Political �0.0007 0.0137 �0.8328Composite �0.0007 0.0103 �0.7032

Iraq Economic 0.0033 0.1117 �0.7442Financial 0.0036 0.1391 1.6272Political 0.0030 0.0558 0.8633Composite 0.0033 0.0486 �0.6748

Malaysia Economic 0.0000 0.0229 �0.6627Financial 0.0009 0.0255 �4.9462Political �0.0003 0.0130 �0.0264Composite 0.0000 0.0118 �1.7884

Mexico Economic 0.0019 0.0359 0.7234Financial 0.0023 0.0312 �1.3627Political 0.0001 0.0160 0.7132Composite 0.0010 0.0160 �0.6843

Romania Economic 0.0007 0.0602 �1.0832Financial 0.0010 0.0688 �2.5329Political 0.0007 0.0189 1.4416Composite 0.0008 0.0207 �0.7155

Saudi Arabia Economic �0.0003 0.0419 0.1291Financial 0.0013 0.0293 0.1001Political 0.0013 0.0266 2.0621Composite 0.0008 0.0203 1.2961

South Africa Economic 0.0000 0.0221 �0.4408Financial 0.0000 0.0278 1.1759Political 0.0001 0.0205 3.1660Composite 0.0000 0.0140 1.6166

Zimbabwe Economic �0.0030 0.0498 �0.6041Financial 0.0010 0.0413 1.7427Political �0.0005 0.0274 0.6096Composite �0.0005 0.0207 �0.4461

44 HOTI AND MCALEER


UNCORRECTE

DPROOF

Table 25. Correlation Coefficients for Risk Returns by Country.

Country Risk returns Economic Financial Political Composite

Albania Economic 1.000 0.077 0.312 0.725Financial 1.000 0.089 0.477Political 1.000 0.749Composite 1.000

Argentina Economic 1.000 0.063 �0.021 0.581Financial 1.000 0.276 0.623Political 1.000 0.675Composite 1.000

Chile Economic 1.000 0.187 0.026 0.725Financial 1.000 0.227 0.592Political 1.000 0.618Composite 1.000

Cuba Economic 1.000 0.108 0.380 0.701Financial 1.000 0.271 0.667Political 1.000 0.751Composite 1.000

Indonesia Economic 1.000 0.124 0.047 0.572Financial 1.000 0.244 0.727Political 1.000 0.649Composite 1.000

Iraq Economic 1.000 �0.056 0.026 0.603Financial 1.000 0.205 0.520Political 1.000 0.653Composite 1.000

Malaysia Economic 1.000 0.161 0.138 0.662Financial 1.000 0.094 0.640Political 1.000 0.641Composite 1.000

Mexico Economic 1.000 0.056 0.188 0.629Financial 1.000 0.286 0.645Political 1.000 0.735Composite 1.000

Romania Economic 1.000 �0.072 �0.068 0.490Financial 1.000 0.017 0.676Political 1.000 0.459Composite 1.000

Saudi Arabia Economic 1.000 0.177 0.000 0.645Financial 1.000 0.289 0.638Political 1.000 0.675Composite 1.000

South Africa Economic 1.000 0.018 �0.035 0.389Financial 1.000 0.159 0.601Political 1.000 0.774Composite 1.000

Zimbabwe Economic 1.000 �0.026 0.043 0.508Financial 1.000 0.052 0.527Political 1.000 0.707Composite 1.000



UNCORRECTE

DPROOF

with the composite risk returns, but not with each other. For eight countries,namely Albania, Argentina, Cuba, Iraq, Mexico, Saudi Arabia, South Africa andZimbabwe, the highest correlation coefficient is between the political and compo-site risk returns. Of these eight countries, the second highest correlation forAlbania, Cuba, Iraq and Saudi Arabia is between economic and composite riskreturn, while for Argentina, Mexico, South Africa and Zimbabwe the secondhighest correlation coefficient is between financial and composite risk returns.For Chile and Malaysia, the highest correlation coefficient is between the economicand composite risk returns, while for Indonesia and Romania the highest correl-ation coefficient is between the financial and composite risk returns.

The risk returns and associated volatilities for the twelve countries are given inFigures 1b–12b. Substantial differences are evident in the risk returns, as well as intheir volatilities. Both Albania and Romania have noticeable outliers for three of thefour risk returns, the exception being political risk returns, for which there is aclustering of volatilities. Argentina has outliers in the case of financial and compositerisk returns, and clustering for the other two risk returns, whereas Chile has cluster-ing in the case of all four risk returns. Outliers are evident in three of the four riskreturns for Cuba, with the exception being economic risk returns, for which thereappears to be little clustering. In the case of all four risk returns for Mexico, outliersseem to be present. With the exception of composite risk returns for Indonesia, andpolitical risk returns forMalaysia, outliers are more obvious than clustering. There isevidence of clustering of volatilities only in the case of political risk returns forMalaysia. Volatilities seem to cluster only for economic risk returns for SaudiArabia, with outliers seeming to dominate the remaining three risk returns. Outliersare also evident for Iraq in the case of financial and political risk returns, but withlittle evidence of clustering of volatilities. South Africa and Zimbabwe displaydifferent patterns. Outliers are present in all four risk returns for South Africa,and clustering for financial risk returns, whereas Zimbabwe has outliers in the caseof economics and political risk returns and clustering for composite risk returns.

6. Concluding Remarks

This paper evaluated the significance of 50 published empirical papers in thecountry risk literature according to established statistical and econometric criteriaused in estimation, evaluation and forecasting. Such an evaluation permits acritical assessment of the relevance and practicality of the economic, financialand political theories pertaining to country risk. Discussion of the empiricalfindings relating to the published studies included descriptions of the countryrisk rating systems by the leading commercial analysts of country risk which wereused, namely Institutional Investor, Euromoney, Moody’s, Standard and Poor’s,International Country Risk Guide, and Political Risk Services.

The rating system of International Country Risk Guide (ICRG), which is theonly risk rating agency to provide detailed and consistent monthly data over anextended period for a large number of countries, was discussed in detail. Acomparison of ICRG country risk ratings, risk returns and associated volatilities

46 HOTI AND MCALEER


UNCORRECTE

DPROOF

was provided for twelve developing countries, representing six geographicregions. The time series data permitted a comparative assessment of the interna-tional country risk ratings, and highlighted the importance of economic, financialand political risk ratings as components of a composite risk rating.

Future research in the area could analyse the monthly time series data for allthe countries covered by the ICRG. The monthly data on country risk ratings andreturns could be used to estimate and test a variety of models, including ARMAmodels of risk returns, and both time-varying conditional volatility and stochasticvolatility models of innovations to risk returns. Univariate and multivariatemodels of volatility could be used to estimate alternative static and dynamicconditional correlation matrices of the innovations to risk returns in order toexamine the direction of any causality in the economic, financial, political andcomposite risk ratings across countries. Hoti et al. (2002) have estimated severalstatic conditional correlation matrices for Australia, Canada, Japan and the USAusing alternative multivariate time-varying conditional volatility models. Thisanalysis could be extended to estimate and test alternative dynamic multivariatetime-varying conditional volatility and stochastic volatility models.

Acknowledgements

The authors wish to thank an editor, two referees and seminar participants at theUniversity of Rome ‘La Sapienza’, the University of Western Australia, and the GlobalOutlook Meeting, Institute of International Affairs, Rome, for helpful comments andsuggestions, and the editor and staff of the International Country Risk Guide for helpfulcorrespondence regarding the data. The first author wishes to acknowledge the financialsupport of an Australian Research Council PhD Scholarship, the C.A. Vargovic MemorialFund at UWA, and an Individual Research Grant from the Faculties of Economics &Commerce, Education and Law at UWA. The second author wishes to acknowledge thefinancial support of an Australian Research Council Discovery Grant.

References

Bourke, P. (1990) Risks in international banking. In P. Bourke and B. Shanmugam, (eds),An Introduction to Bank Lending. Sydney: Addison-Wesley Business Series.

Brewer, T. L. and Rivoli, P. (1990) Politics and perceived country creditworthiness ininternational banking. Journal of Money, Credit and Banking, 22(3): 357–369.

Burton, F. N. and Inoue, H. (1985) An appraisal of the early-warning indicators ofsovereign loan default in country risk evaluation systems. Management InternationalReview, 25(1): 45–56.

Cosset, J. C. and Roy, J. (1991) The determinants of country risk ratings. Journal ofInternational Business Studies, 22(1): 135–142.

Eaton, J. and Taylor, L. (1986) Developing country finance and debt. Journal of DevelopmentEconomics, 22: 209–265.

Ghose, T. K. (1988) How to analyse country risk. Asian Finance, October, 61–63.Hayes, N. L. (1998) Country risk revisited. Journal of Lending and Credit Risk Management,

80(5): 61.Hoti, S., Chan, F. and McAleer, M. (2002) Structure and asymptotic theory for multivariate

asymmetric volatility: Empirical evidence for country risk ratings, paper presented at theAustralasian Meeting of the Econometric Society, Brisbane, Australia, July.



UNCORRECTE

DPROOF

Howell, L. D. (2001) The Handbook of Country and Political Risk Analysis, 3rd edn.New York: PRS Group.

Juttner, D. J. (1995) Risk premia in foreign exchange and interest rates. In InternationalFinance and Global Investments, 3rd edn. Melbourne: Longman, chapter 16.

Krayenbuehl, T. E. (1985) Country Risk: Assessment and Monitoring, Toronto: LexingtonBooks.

McAleer, M. (1994) Sherlock Holmes and the search for truth: a diagnostic tale. Journalof Economic Surveys, 8: 317–370. Reprinted in L. Oxley et al. (eds), Surveys inEconometrics, Oxford: Blackwell, pp. 91–138.

Nagy, P. J. (1988) Country Risk: How to Assess, Quantify, and Monitor It. London:Euromoney Publications.

Overholt, W. H. (1982) Political Risk. London: Euromoney Publications.Ramcharran, H. (1999) International bank lending to developing countries: an

empirical analysis of the impact of country risk. Multinational Business Review, 7(1):83–91.

Rockerbie, D. W. (1993) Explaining interest spreads on sovereign Eurodollar loans: LDCsversus DCs, 1978–84. Applied Economics, 25: 609–616.

Saini, K. G. and Bates, P. S. (1984) A survey of the quantitative approaches to country riskanalysis. Journal of Banking and Finance, 8: 341–356.

Shanmugam, B. (1990) Evaluation of political risk. In P. Bourke and B. Shanmugam, AnIntroduction to Bank Lending. Sydney: Addison-Wesley Business Series.

APPENDIX

Published Empirical Studies on Country Risk

Abdullah, F. A. (1985) Development of an advanced warning indicator of external debtservicing vulnerability. Journal of International Business Studies, 16(3): 135–141.

Backer, A. (1992) Country balance sheet data vs. traditional macro variables in a logitmodel to predict debt rescheduling. Economics Letters, 38: 207–212.

Balkan, E. M. (1992) Political instability, country risk and probability of default. AppliedEconomics, 24(9): 999–1008.

de Bondt, G. J. and Winder, C. C. A. (1996) Countries’ creditworthiness: an indicator froma probit analysis. De Economist, 144(4): 617–633.

Brewer, T. L. and Rivoli, P. (1990) Politics and perceived country creditworthiness ininternational banking. Journal of Money, Credit and Banking, 22(3): 357–369.

Burton, F. N. and Inoue, H. (1985) An appraisal of the early-warning indicators ofsovereign loan default in country risk evaluation systems. Management InternationalReview, 25(1): 45–56.

Cantor, R. and Packer, F. (1996) Determinants and impact of sovereign credit ratings.FRBNY Economic Policy Review, October, 37–53.

Chattopadhyay, S. P. (1997) Neural network approach for assessing country risk forforeign investment. International Journal of Management, 14(2): 159–167.

Citron, J. T. and Nickelsburg, G. (1987) Country risk and political instability. Journal ofDevelopment Economics, 25: 385–392.

Cooper, J. C. B. (1999) Artificial neural networks versus multivariate statistics: an applica-tion from economics. Journal of Applied Statistics, 26(8): 909–921.

Cosset, J. C. and Roy, J. (1991) The determinants of country risk ratings. Journal ofInternational Business Studies, 22(1): 135–142.

Doumpos, M. and Zopounidis, C. (2001) Assessing financial risks using a multicriteriasorting procedure: the case of country risk assessment. Omega, 29(1): 97–109.

Easton, S. T. and Rockerbie, D. W. (1999) What’s in a default? Lending to LDCs in theface of default risk. Journal of Development Economics, 58: 319–332.

48 HOTI AND MCALEER


UNCORRECTE

DPROOF

Eaton, J. and Gersovitz, M. (1980) LDC participation in international financial markets:debt and reserves. Journal of Development of Economics, 7: 3–21.

Eaton, J. and Gersovitz, M. (1981) Debt with potential repudiation: theoretical andempirical analysis. Review of Economic Studies, 48: 289–309.

Edwards, S. (1984) LDC foreign borrowing and default risk: an empirical investigation,1976–80. American Economic Review, 74(4): 726–734.

Erb, C. B., Harvey, C. R. and Viskanta, T. E. (1996) Political risk, economic risk andfinancial risk. Financial Analysts Journal, November/December, 29–46.

Feder, G. and Just, R. E. (1977) A study of debt servicing capacity applying logit analysis.Journal of Development Economics, 4: 25–38.

Feder, G., Just, R. E. and Ross, K. (1981) Projecting debt servicing capacity ofdeveloping countries. Journal of Financial and Quantitative Analysis, 16(5): 651–669.

Feder, G. and Uy, L. V. (1985) The determinants of international creditworthiness andtheir policy responses. Journal of Policy Modelling, 7(1): 133–156.

Frank, C. R. and Cline, W. R. (1971) Measurement of debt servicing capacity: an applicationof discriminant analysis. Journal of International Economics, 1: 327–344.

de Haan, J., Siermann, C. L. J. and van Lubek, E. (1997) Political instability and countryrisk: new evidence. Applied Economics Letters, 703–707.

Hajivassiliou, V. A. (1987) The external debt repayments problems of LDC’s: an econometricmodel based on panel data. Journal of Econometrics, 36: 205–230.

Haque, N. U., Kumar, M. S. Mark, N. and Mathieson, D. J. (1996) The economic contentof indicators of developing country creditworthiness. International Monetary FundStaff Papers, 43(4): 688–724.

Hernandez-Trillo, F. (1995) A model-based estimation of the probability of default insovereign credit markets. Journal of Development Economics, 46: 163–179.

Kharas, H. (1984) The long–run creditworthiness of developing countries: theory andpractice. Quarterly Journal of Economics, August, 415–439.

Kugler, P. (1984) Economic indicators and debt reschedulings 1977–1981: empirical resultsfrom a logit model. Rivista Internazionale di Scienze Economiche e Commerciali,10(11): 992–1005.

Kutty, G. (1990) Logistic regression and probability of default of developing countriesdebt. Applied Economics, 21: 1649–1660.

Lanoie, P. and Lemarbre, S. (1996) Three approaches to predict the timing and quantity ofLDC debt rescheduling. Applied Economics, 28: 241–246.

Lee, B. C. and Powell, J. G. (1995) A behavioural approach to sovereign debt reschedul-ings. Applied Economics Letters, 2: 64–66.

Lee, S. H. (1991a) Ability and willingness to service debt as explanation for commercialand official rescheduling cases. Journal of Banking and Finance, 15: 5–27.

Lee, S. H. (1991b) Using terms of rescheduling as proxy for partial reneging of LDC’s debtin a test of willingness-to-pay model. Journal of International Money and Finance, 10:457–477.

Lee, S. H. (1993a) Relative importance of political instability and economic variables onperceived country creditworthiness. Journal of International Business Studies, FourthQuarter, 801–812.

Lee, S. H. (1993b) Are the credit ratings assigned by bankers based on the willingness ofLDC borrowers to repay? Journal of Development Economics, 40: 349–359.

Lloyd-Ellis, H., McKenzie, G. W. and Thomas, S. H. (1989) Using country balance sheetdata to predict debt rescheduling. Economics Letters, 31: 173–177.

Lloyd-Ellis, H., McKenzie, G. W. and Thomas, S. H. (1990) Predicting the quantity ofLDC debt rescheduling. Economics Letters, 32: 67–73.

Mascarenhas, B. and Sand, O. C. (1989) Combination of forecasts in the internationalcontext: predicting debt rescheduling. Journal of International Business Studies, 20:539–552.



UNCORRECTE

DPROOF

Moon, C. G. and Stotsky, J. G. (1993) Testing the differences between the determinantsof Moody’s and Standard and Poor’s ratings: an application of smooth simulatedmaximum likelihood estimation. Journal of Applied Econometrics, 8: 51–69.

Morgan, J. B. (1986) A new look at debt rescheduling indicators and models. Journal ofInternational Business Studies, 17: 37–54.

Odedokun, M. O. (1995) Analysis of probability of external debt rescheduling in Sub-Saharan Africa. Scottish Journal of Political Economy, 42(1): 82–98.

Oetzel, J. M., Bettis, R. A. and Zenner, M. (2001) Country risk measures: how risky arethey? Journal of World Business, 36(2): 128–145.

Oral, M., Kettani, O. Coset, J. C. and Daouas, M. (1992) An estimation model for countryrisk. International Journal of Forecasting, 8: 583–593.

Rahnama-Moghadam, M., Samavati, H. and Haber, L. J. (1991) The determinants of debtrescheduling: the case of Latin America. Southern Economic Journal, 58: 510–517.

Rahnama-Moghadam, M. (1995) Debt rescheduling in less-developed countries: world orregional crisis? Journal of Developing Area, 30: 11–22.

Ramcharran, H. (1999) The determinants of secondary market prices for developingcountry loans: the impact of country risk. Global Finance Journal, 10(2): 173–186.

Rivoli, P. and Brewer, L. T. (1997) Political instability and country risk. Global FinanceJournal, 8(2): 309–321.

Schmidt, R. (1984) Early warning of debt rescheduling. Journal of Banking and Finance, 8:357–370.

Scholtens, B. (1999) On the comovement of bond yield spreads and country risk ratings.Journal of Fixed Income, 8(4): 99–103.

Somerville, R. A. and Taffler, R. J. (1995) Banker judgement versus formal forecastingmodels: the case of country risk assessment. Journal of Banking and Finance, 19: 281–297.

Taffler, R. J. and Abassi, B. (1984) Country risk: a model for predicting debt servicingproblems in developing countries. Journal of the Royal Statistical Society Series A,147(4): 541–568.

50 HOTI AND MCALEER


Author Query Form

Journal: Journal of Economic Surveys

Article : 001

Dear Author,

During the copy-editing of your paper, the following queries arose. Please respond to these by marking up yourproofs with the necessary changes/additions. Please write your answers on the query sheet if there is insufficient spaceon the page proofs. Please write clearly and follow the conventions shown on the attached corrections sheet. Ifreturning the proof by fax do not write too close to the paper’s edge. Please remember that illegible mark-ups maydelay publication.Many thanks for your assistance.

QueryRefs.

Query Remarks

1 Au: Please provide the keywords.

2 Au: ‘Sephton and Cochrane, 1990’ has not been included in thereference list. Please check.

3 Au: Sentence has been rephrased. Please check if it is OK.

4 Au: Sentence has been rephrased. Please check if it is OK.

5 Au: ‘Moore & Cullen 1985’ has been changed to ‘Moore andCullen 1995’, so that this citation matches with that in thereference list. Please check if it is OK.

6 Au: ‘Chen et al. 1990’ has not been included in the reference list.Please check.

7 Au: the letter ’a’ has been added to the year



10 Au: ‘Chen et al. 1990’ has not been included in the reference list.Please check.

11 Au: ‘Brunetti & Gilbert 1997’ has been changed to ‘Brunetti andGilbert 1995’, so that this citation matches with that in thereference list. Please check if it is OK.

QueryRefs.

Query Remarks

12 Au: ‘Hansen & Hoderick 1980’ has been changed to ‘Hansen andHodrick 1980’, so that this citation matches with that in thereference list. Please check if it is OK.





17 Au: ‘Chang et al., 1990’ has not been cited in the text. Pleasecheck.

18 Au: Please provide the city of the publisher.

19 Au: Please provide the volume number.

20 Au: Please provide the full form of ‘NCE’.

21 Au: ‘Agbevebe 1992’ has been changed to ‘Agbeyegbe 1992’, sothat this citation matches with that in the reference list. Pleasecheck if it is OK.

22 Au: ‘Krehbiel & Adkms 1993’ has been changed to ‘Krehbiel andAdkins 1993’, so that this citation matches with that in thereference list. Please check if it is OK.

23 Au: Please clarify whether ‘rational expectations no risk premium’can be ‘rational expectations’ no-risk premium’. Or, please makethe necessary change.

an empirical assessment of country risk ratings and ... · inoue, 1985). country risk assessment...

Documents